• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

GAMMA 30 experiement

Hmm gotta' learn a new language I guess (I won't though), this is the info on the speaker in the last pic above.

index.php


index.php


The relevant thread:
http://translate.google.com/transla...u=http://penkiller.com/index.php?topic=637.15
 
So I'm having a bit of trouble finding out if I have horn shouting or not, I probably do.

Some recordings sound great, quiet or loud (though never super loud, and I don't mind), all is good.

Other recordings sound fine a low-moderate volume, but get sibilant and shouty at louder volumes.
Those recordings are edgy on the K240DF's too, so it's not entirely the speakers fault.

The speakers must be quite sensitive to sibilance though.

Why don't I just blame the horns? they are crossed over low, I should expect horn shouting if I push them.

Because there are two long hard plastic reflex tubes in both of the boxes, right next to the woofers.
Both are stuffed with closed cell foam, so they are not even in use, but do they cause resonances or nasty reflections inside the box, that make their way out through the driver? I donno...
Also, there is a stuffed horn and a box for midwoofer in there too, one of the boxes is too leaky, and they are sparsely lined with fiberglass for BR duty.

And I still haven't put phase plugs on the woofers.
Eliminating their inner cavity by removing the dustcap did wonders, phase plugs should decrease upper end distortion a bit, and improve the dispersion pattern.
Who knows what the dispersion pattern is like without the dust cap, maybe that is contributing to what I'm hearing.

Oh well, if I do the same mods to a new pair of LA1231's, like the phase plugged one above, I should be able to cross at 4KHz no problem, the designer of those, claims smooth response up to 7-8Khz, plenty of room to play with.

I should leave the crossover alone for now, fix the box, make phase plugs, try them out, buy new LA1231's, modify them, and see where I am.
 
I have now finished the first BETA version of my crossover, by calculating the Q value of the original low pass filter, and modifying my filter to have the same Q value, but crossed lower (slightly steeper as the high pass filter is also slightly steeper then then original).
It turned out all I had to do was put my 4.7uF caps in parallel with the woofers along with the 6.8uF ones.
So I didn't buy a single ClarityCap too many!

I have listened to this, constantly for days, I can't detect anything wrong with it, the sibilance problem is gone, and horn shouting isn't a problem unless I play louder then I want to.

ec0a669a_GAMMAmkII.png
 
Last edited:
Now that I take an even closer listen, some S sounds sound wrong, there might be a slight hump in the 2000-2500Hz range.

I'm going to try some higher DCR coils for the tweeter, still 0.3mH.
 
I decided to try and mimic the original crossover more closely, by putting the 4.7uF cap back in to the highpass filter, and the 3.9uF cap in parallel with the woofer.

This makes a lowpass filter slope with a Q value almost identical to the original crossover, since the highpass filter is default, the slopes should at least maybe possibly fit, as long as they cross at the right frequency.

And yes indeed, "S" sounds sound better, and I'm really starting to understand what phase distortion does.

If I don't reverse wire the tweeters, sound appears to come from them.
If I alter the crossover slopes too much compared to original, "phantom" speaker effect between the speakers, seems a bit diffuse.

Now it's good though, and sound in the 2-3KHz area sound nice and focused, a small but very noticeable difference, once you discover how to listen for it, it's not obvious at first.

I'll probably get horn shouting a low volume now though, will have to check that tomorrow.

I think the reason why I have had to change the crossover so much compared to the original, has to do with driver placement and coil DCR values.
In the original GAMMA 30 design, the woofer is at floor level, with the tweeter above it, both in the same box, the crossover uses directivity to assist the crossover.

Since I place the woofer high with the tweeter on top of the box, I have had to cross the woofer lower.

Also the inductors I use certainly have lower DCR then the originals, I don't how much lower, this will have had some sort of effect as well.

So this would be BETA v1.01 :D
15d1f0b9_GAMMAmkII.png


I'm still going to try higher DCR coils on the tweeters though.
 
Hmm I've been at this too long, now ALL sound systems appear to have too forward S, T, F sounds and so on.

I need to get back in to a state where I can hear the music, rather then analyze it, I have made my self too sensitive in the 2-3KHz area.

BUT, since my headhpones make the same kind of S sounds, I must have done alright, also the phase thing matches up.

Play loud? I don't know, it all sounds too S'y, now I have to force my self to stop, I need a long evaluation period, and my hearing needs to normalize.
 
Nope, my speakers don't sound S'y at all, my ears had just adapted to the job I asked them to do.

Still waiting for some higher DCR 0.3mH coils, at this point I have to use music to get to the final voicing.
I'm so close that tone sweeps only reveal tiny "problems" that are hard to replicate, I think that is because of directivity issues.

Phase plugs are making their way to the very top of the list, but the only person I know who can make some, only works with metals (aluminum).
Wood is supposed to be better, but I think that if I coat the aluminum with something, it should be fine.

The real problem is the shape, I have to try and copy the shape of the ones in the speaker I pictured earlier and hope for the best.
 
New coils have been put in the highpass filter, same 0.3mH size, but 0.43ohm DCR.

The result is a warmer sound, recordings with problematic "S" sounds are now sound more forgiving.

There is a very slight dip in the 1.6-1.8KHz range, followed by a slight wide hump, peaking at 2KHz and leveling off somewhere in the 2.5KHz range.

These problems are very slight though, and listening tests are very satisfying.

I think I'll try and swap the 3.9uF used in the lowpass filter, and the 4.7uF from the highpass again once I feel like tinkering some more.
 
I think I might have to settle for a compromise, the imaging and focus I'm getting right now is spot on!

What I can say for sure is that 1.5KHz is louder then 1.6Khz, there is a sudden rise at 1.9KHz, peaking at 2KHz.
2KHz is louder then 3KHz, 1.5KHz is as loud as 3KHz.

If this is the price of this kind of imaging, then so be it.

All very slight variations, and remember, it is very difficult to compare amplitude, when the tone has changed a lot.
Using ears for tone sweeps is effective at spotting peaks or dips, but there is a limit to how shallow slopes you can hear.
Tone sweeps don't reveal phase issues well.

Best thing to do at this point, must be to let my ears decide.
 
The frequency range anomalies I described above are caused by one of the speakers, the other one sounds flat in the crossover range.

I swapped tweeters to see if the problem followed one of them, but no it made no difference.

Maybe I got poorly matched crossover components, maybe the woofers are worn down, the affected speaker has a very leaky box, or maybe it's a room problem.

Edit: Tolerances for the caps is 3%, 5% for the coils, not big enough to cause such a difference, must be a room, woofer or box problem.

d4ebc898_GAMMAmkII.png
 
Last edited:
I have tried removing the tweeters from the equation, and listened to box, lowpass and woofers only.

The leaky speaker definitely plays louder at 2KHz then the fairly well sealed one.

So the problem must be that the box is leaky, or maybe the voice coil in the woofer is damaged, giving it a lower impedance then it should have.

I really don't think this is a problem with the crossover, I just hope that the speaker that sounds good, is the one that isn't "broken" :S
 
I had a feeling that maybe one woofer was more efficient then the other, since they have AlNiCo magnets that is a real possibility.

But I can only detect higher amplitude in the "problem" speaker above 1KHz going up to around 3.2KHz.

It really smells like a damaged voice coil, reading up on speaker boxes have told me that the box doesn't cause these sorts of things.

I'll have to switch the speakers around, to see of the problem follows the speaker, or stays at the same side.
 
Last edited:
The problem follows the speaker, I guess it's a woofer problem.
The woofer that is behaving nice, is the one that originally had a dislocated magnet, that I moved back in to place, that sits in a fairly well sealed box, I'm pretty sure it's surrounds are air tight, since the box is (driver feels bouncy).
The other one, might still have a leak in it's surrounds, I know fixing those made a substantial improvement, it's box needs work though, that's why it leaks.

I have tested the magnets in the woofers, using a screwdriver (handle without sharp edges, that isn't so strongly magnetic that it could cause damage), both magnets feel equally strong.

I think this is a voice coil problem, I should order new LA1231's sooner then later.
 
OK today I'm in a tinkering mood, and I'll try a new direction, as I have quite a collection of coils and good caps by now.

No crossover is the best crossover, so how little crossover can I get away with? does my BK3013A's really need a filter at all? do I have the components I need to make a highpass, preferably 2nd order, to integrate the tweeter with the woofer, where the woofer has no filter at all?

The target crossover range will be 4-6KHz.

These are the questions that are bugging me today, and I'll have to see how simple I can make it, it's time to see if I can apply what I have learned about crossovers.

First things first, I have to test the woofers without crossover, and see if they sound the same.
 
I just made this filter, and installed it in one the speakers, very exciting, seems good enough for me to want to do it to the other speaker as well.

05bb7b0a_GAMMAmkIII.png
 
Last edited:
It's a success!!!

The channel imbalance I had, might have been due to crossover component variations, I'm not so sure though.
Maybe it was rather poor solder work as my soldering iron is JUNK!

Anyway, I took the time to get all the joints right, woofer running full range, minimal network (for a ribbon) on the tweeter, with very low DCR, so dampening factor should be excellent throughout.

They sure are fast!

Tone sweeps did not reveal any problems at this time.

Power handling should be greatly increased as well, crossover frequency is 5KHz with a 12" woofer :D

I definitely have a directivity problem on my hands though!
 
Last edited:
Oh almost forgot, the woofer that had a dislocated magnet, I readjusted it, now that the dustcap is off, it's much easier.
That could be what solved (at least I think so at this point) the channel imbalance.
 
LOOOOOL oh yeah!

Just made as quick attempt at playing my new crossover loudly, and damn they play loud, and they don't start sounding harsh or sharp or anything.
They basically just play louder and louder and louder and it's pretty damn awesome.

I'm not sure where the limit is, at this point, the person living below me will want me dead if I try and find out, so they definitely play LOUD.

Probably not party speakers, but it just might be doable, I'm pretty sure they won't blast a golf field though.

Edit: hmm, my speakers really aren't GAMMA 30's anymore, the GAMMA 30 is a 50 liter BR speaker with woofer at floor level crossed in the 2-3KHz region for low SPL.
What I have now is crossed at 5KHz for high SPL, using a completely different kind of crossover and it's all put in to a sealed box.

I guess thye could be called a GAMMA 30 mkIII, but I think they are just DIY GAMMA 2-ways now.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom