2 way crossover design help

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Whatever works. I won't argue with results.

I had not thought of making them shallower, but had thought about making them shorter. Just worked up passive BSC and 2nd order crossover to add the XT25's. I think the BSC will balance the mid base much better and I won't have to adjust the size. I hope the second order helps on the breakup.
 
I Like it

Need to finish the second one...but so far so good.
 

Attachments

  • V3.jpg
    V3.jpg
    278.1 KB · Views: 124
  • V 4.jpg
    V 4.jpg
    311.7 KB · Views: 116
What got this started is the box tuning was to high, to lower the tuning the slot/ports would have been the size of straws........:-/

In that respect - mission accomplished; "low" tuning with huge ports. This should be be good; can't wait to finish up.

Another pair right behind....Aye Carumba!!
 
Tvr,

You the nail on the head a couple of times, the full range boxes do chuff a bit, not so much you can hear it over the music.

They are sounding nice, light and quick. I did pull out huge pieces of damping material (since the last time I had them together) and I like them better than before.

The midrange breakup...maybe that's why they need hours of break-in-time.

Let me know when you get you go mobile with mic, etc.




I just finished doing my own TS measurements on the Fostex. Qes came in at .56 and Qms at 7.35. Fs was 72 with VAS at 7.3L. Not as far off as I have seen. Usually it is VAS they lie like a dog about but this is only a couple of percent. The frequency response on the spec sheet, though very small, is closet to what I measure than the model. Of course, it does not show baffle step but the bad breakup at 6800 is there for all to see. I can't see how anyone could consider this driver a full range. A 200 to 2K is more like it. Even that, the harmonic distortion is about 3.5% This means the 11L boxes are tuned at 60 Hz. I think I would try them at 75 Hz. I am going to play with the electronic LR-4 crossover between 3 and 4K to see if it handles the resonance OK. Not as high as I want, but well into the easier range for a dome tweeter. Too low for a ribbon. I can try LR6, but that does force one to stay electronic.
 
Last edited:
Chuff - Chuff Eliminated!

Well the chuffing is gone now....built new boxes with 1/16" WIDER slots, also rounded over the inter panels (inside) with round-over bit.

Going to tune by ear A/B, comparing with the other boxes;it's tuned right around Fostex specs (57hz) with the untrimmed filler pieces. I need to be carful not to trim them too much; or will be gluing up more.

Will of course use generator also.

btw, let me know if you need any of these prototypes..........




This is a recipe for chuff-chuff noises and also requires differenced port length calculations. When feeding old British sports cars, the correctly designed "velocity stacks" meant as much ad 10% more power. This is fluid dynamics. The math is way the heck beyond me. I use a sim to get close and tune to measure. I would be concerned with the above picture that you will get wind noises by the relative small opening on the box end compared to the mass of air in the port.
 

Attachments

  • DSCN0225.JPG
    DSCN0225.JPG
    326.8 KB · Views: 89
We can easily measure the actual box tuning with just my WT-2. Then compare what you hear to what we measure. Sometimes surprisingly good things come out of that, and occasionally hints when it does not work well.

While I am in between primer coats, I'll work on getting my portable rig set up.
 
diyAudio Moderator
Joined 2008
Paid Member
I would be concerned with the above picture that you will get wind noises by the relative small opening on the box end compared to the mass of air in the port.

The slot port reminds me of those pedestal room fans with the narrow ring slot but no blades. It seems like an opportunity to optimise the flow which was done with an aerofoil at the opening. The small conventional fan in the base of these with its narrow diameter can be quite noisy.
 
Looking for a tweeter for this box

We had a little listening get together yesterday. The small dark wood covered boxes (pic in post #1) were holding their own. Lots of solid bass, imaging is very good (which I always thought) but it was good to hear it from somebody else. We tried several pairs of speakers; this was the runner up :note:

So the second part of this post. Looking for a tweeter for this box, it sounds pretty good for a full range, lacking top end crack, e.g. can't hear wood hitting cymbals. The imaging is impressive, just no high end :grumpy: I was thinking about a small tweeter to handle 4,000 >

Vifa XT25SC90-04 1" Dual Ring Radiator Tweeter 264-1014

This is the woofer in the box:

Tang Band W5-1611SAF 5" Full Range Speaker 264-918

btw, getting some really nice lows out of this box, really low.
 

Attachments

  • Tweeter I.jpg
    Tweeter I.jpg
    44.8 KB · Views: 121
Last edited:
Finished Product

Here's a pic of the TV/speakers. I tried to make it easy for the wife to use so using the TV's mini 1/8" R/L out, right into a amplifier, however losing a lot of the high end.

I guess I could bypass the 0.1 with a clip lead and see if there is significant increase in the treble.

Again - Thanks for the help!

[need to make mini stands for the speakers]


I'd GUESS looking at Thomas Akerlund's circuit there, that the 4 ohm peerless has Le of 0.55mH...

So this looks reasonable to me as a start:

An externally hosted image should be here but it was not working when we last tested it.


The 0.1 ohm resistor is adjustable for tweeter level. The 0.55mH coil will mean there isn't much bafflestep correction, so these are best placed by a wall. Think that agrees with Jerome69's ideas too. Really can't say how bad the woofer peaks are going to sound, or exactly how to tame them. But this circuit is non-critical and has good phase alignment from 2kHz to 10kHz.
 

Attachments

  • Temp 12.jpg
    Temp 12.jpg
    80.8 KB · Views: 100
Try Everything!

I spent most of the day on this yesterday -

There was one mistake I found early, the set-top-box was set to "heavy" which is like a "loud" button; turned that off.

I gave up on the digital (HDMI) hook up (for sound), I was going thru the TV's D/A converter for a preamp, it was adding color to the sound.

The better hook up: was Analog - Set-Top-Box to amp (only). The STB has a variable out, so can adjust level from easy chair - but there is no switching.

The tweeter needed a little help, added small caps until the tweeter sounded balanced (I didn't think the middle was a issue) I went from a 3.9uf to 5.9uf (as shown).

The xo circuit works very well, the imaging is very good, with a stereophonic sound.

:D
 

Attachments

  • Temp14.jpg
    Temp14.jpg
    28 KB · Views: 83
Last edited:
This is all great stuff, ODougbo! I have in mind a similar project with some small 4ohm polycones I have lying about.

I get in terrible trouble for saying this stuff is all quite straightforward, but it's interesting that you have ended up with something not a million miles from the Epos ELS-3 design. Naturally, 2nd order tweeter is better with time-alignment, 3rd order on flat baffles. We don't find tweeter phase reversal and time-alignment too difficult either, eh?

Up the diy-ers for MUSICAL speakers! :cool:
 

Attachments

  • Epos_ELS-3_Loudspeaker.JPG
    Epos_ELS-3_Loudspeaker.JPG
    23.6 KB · Views: 42
  • Epos ELS-3 Crossover.JPG
    Epos ELS-3 Crossover.JPG
    15.8 KB · Views: 44
  • Epos_ELS-3_FR.JPG
    Epos_ELS-3_FR.JPG
    25.8 KB · Views: 35
  • Epos_4ohm_130mm.JPG
    Epos_4ohm_130mm.JPG
    17.2 KB · Views: 39
Last edited:
I'm really enjoying them (been on for hours now)....They never sounded better.

The plywood/angled/sand boxes may be helping

btw, only using 1.5oz of speaker wool.

Over the last several weeks, been trying different caps for tweeter:

3.9uf - Highs were a bit hard to hear
6.0uf - Too large (but some songs sounded great!)
4.7uf - Seems to be in the right range - a nice balance
 
Did you ever find a PDF with frequency response for the Peerless 832591 4 ohm driver, ODougbo?

Reason I ask is the Peerless 832592 8 ohm version is just crying out for a notch filter at 4kHz. :)

15uF and about 0.1 mH and 1.5 ohm in series on the bass shunt for a 4 ohm speaker...:cool:

It's just nagging away at me. Haven't calculated the coil exactly, but it's in the right ballpark. :eek:
 

Attachments

  • Peerless_832592.PNG
    Peerless_832592.PNG
    74.2 KB · Views: 31
Never found the PDF - there must be one out there. I'll email Madisound; this was a popular woofer years ago, there must be documentation.

My speaker buddy is set up with ARTA, I'll see if I can twist his arm and see if he'll measure them.

[btw, the woofers have layer of silicon and dust caps were removed]

[btw II, I had some the 8ohm (832592) didn't have the same punch as the 4ohm]
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.