• 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

Coils installed and we are definitely getting close now, I still think I need to put a bigger cap on the low pass filter, I don't have a suitable one at this time.
I might try adding a 2.2uF or a 3.3uF cap in paralel with the 6.8 though, just to see what would happen.

But now I wanna' listen to music!

Edit: I think the crossover point is somewhere around 2500Hz now, maybe even a bit lower, making this configuration very similar to the DECCA Kardioid.

Edit2: I also think that I should probably try and adapt the high pass filter next, I'm asking a lot of my VLD's at this point, they prefer crossing over at 3500Hz.
This won't be easy, but I have a few caps and a coil I can unwind to see how a smaller coil would affect things.
 
Last edited:
I think I'll proceed by putting a 3.3uF cap in place of the 4.7uF in the high pass filter, and replace the coil with one of my spares, which I'll unwind till it looks smaller then 0.3mH, and test it out.

I'll unwind the coil little by little as I test, so that I can follow the changes.

Since dealing with the high pass filter is difficult, this is all I plan on doing to it, then I go back and focus on the low pass filter, if need be :D
 
That test ended in failure, it just ended up making the tweeters more quiet, but the same peaks in the 2000-3000Hz area remained.

I'm a bit at a loot atm. I'll revert the changes, maybe try and remove the 1.5uF cap on the low pass filter, who knows what might happen.
 
Hmm I was using 0.7mm thread in the coil I was using for testing before, so it reduced the sensitivity off the tweeter quite a bit, maybe the trick is to find a coil with the right DCR.

I have a 0.47mH 1.2mm coil, maybe that could be adapted to work, if only I could measure them, time for research.

Edit: I don't have the tools for that.
 
Remove the 1.5uF cap fomr the low pass filter has concentrated the peaks to around 2600Hz and 3100Hz, very narrow peaks.

Edit: So what I need to figure out is, how to roll off the ribbon a little earlier, but with a similar slope, and to reduce it's efficiency slightly.

I guess I should try a thinner coil, of a slightly lower value, like 0.27mH, and maybe a smaller cap in place of the 3.3uF, could be a 3uF...

Edit2: I am so very close, so very very close :D
 
Last edited:
Ok, the 1.5uF cap has been put back in the low pass filter, a 1.2mm coil of largely unknown value has been put in the high pass filter, my guess is that it's somewhere between 0.25mH and 0.30mH.

Only one peak remains around 2600Hz, it's a fairly bad one though.

Edit: Having update the other speaker to the new crossover config, it's obvious that it's difficult to unwind two coils and keep the same values for both.

Edit2: I'm not getting anywhere by modifying the highpass filter.
 
Last edited:
I think I'll get even bigger coils for the low pass filter.
The ribbons may be pushed quite a bit lower then they are designed for, and I'm not sure if the unpleasant sound I get when playing music at loud volumes, is only due to the forward 2000-3000Hz area with it's peaks just above 3000Hz, or my ribbons have begun "shouting" because they are distorting.

On the other hand, all testing I have done on the ribbons, when the bass was not involved, has shown absolutely no signs of weakness.

So I think I'll get a 0.7mH coil for the low pass filter, sounds like I'm repeating my self a little at this point, maybe I should get some 0.8mH ones while I'm at it...

Edit: unpleasant sound on some recordings I might add.
 
Last edited:
I put the 1.5uF cap in parallel with the 4.7uF cap in the highpass filter, it seems to have smoothed the 2000-3000 range out, at low volumes I have hard time detecting any peaks or humps below 3000Hz.
There is a bit at 3100Hz I tihnk, tomorrow I can turn up the volume and find out for sure.

Edit: Something I have learned for sure is it's good to have some small high quality caps around.
 
Last edited:
Eureka!! I know I can suppress the 3100Hz peak, by putting a 1.5uF cap in parallel with the 6.8uF cap in the lowpass filter.

But I only have two of those caps, tomorrow I'll try putting both of those in one speaker, to see if the result is a predicted, this might boost the 2600Hz range a little though.
 
I have some Ero MKT 1822 6.8uF caps, they are supposed to be good for tweeters, so I could put them on the highpass filter, in stead of the 4.7 + 1.5 combo, and then put the 1.5uF caps in the bass!!!

The sun will be up before I get to bed, again...
 
That turned out to be yet another step in the right direction, the results where exactly as predicted, I may have a gentle hump around 2600Hz, doesn't seem to be a problem atm.
 
I Built some ribbons yrs ago in an Audio Amateur article and the impedance was 1 ohm with low output. What's your ribbon impedance?
 
What would you use for driving such ribbons? they would burn most amps out there.

I don't really know the impedance of my tweeters, it's supposed to be around 8 Ohms, mine are probably a bit lower.
 
A bit more about the tweeters.

Their efficiency is probably somewhere around 96dB/w (with horn, probably ~86dB/w without), their frequency response is flat from 3000Hz and up (with horn).
Power handling is a guess, but maybe 70-80w, specs state 60w@4000Hz for the VLD-13.

Edit: as long as the amp isn't clipping or has a lot of DC on the output, these ribbons can play very loud.
 
Last edited:
More testing has shown that the crossover config I came up with yesterday, has it's own problems, the hump at 2600Hz and a smaller hump at 2800Hz.

So I think the way to go is to get a bigger coil for the lowpass filter, the humps I'm getting are probably a result of too steep crossovers using Chebyshev filters.
I need more gradual curves, with no hump right before the crossover point.
 
A few old GAMMA VLD-12/13 measurements:

Without horn:
jordanowvld13lf4.jpg


With horn:
b55ffa57_VLD13.jpeg
 
I found these old measurements of the BK3013A, someone thought they applied to the LA1231 as well though.
Maybe they did, but there are differences between this graph and the recent measurements of the LA1231.
Maybe the changes didn't happen as the driver changed name, but at some later point in time.

See those peaks between 2500Hz and 4000Hz, the two first of those correlate very well with the two peaks I have been fighting, the LA1231 does not have those.

It seems everyone around the web have been mistaking, the BK3013A and the LA1231 do have more differences then just power handling (due to an upgraded voice coil in the LA1231, as far as I have been able to find out).
The BK3013A is rated at 40w here, but the LA1231 is rated at 100w.
The recent ferrite version seems to have lower efficiency then the old AlNiCo one, but hifikit claims it has 94db/w efficiency.

This could also explain why my crossover has needed so much tinkering, maybe the impedance is different between the two.
The impedance curve of the LA1231 suggests a impedance above 14 ohms, I can't read it well, maybe s high as 20 Ohms.

It seems I have to get a pair of LA1231's earlier then I expected, if they turn out to be too inefficient, I could look in to swapping the magnets.

GAMMA BK3013A:
02b95a83_LA1231.jpeg


GAMMA LA1231:
x675


x675
 
A few pics of the magnets, looks interchangeable to me:

Edit: I can't find a pic of the LA1231 version they have at hifikit, at least not with the barrel off.

BK3013A:
339c76f8_BK3013A-2.jpeg


LA1231 (AlNiCo):
PICT2371.jpg


Vissokogovoriteli variant (that impossible name is the factory they come from):
attachment.php
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom