Page 1 of 1

Restacking lams of speakerformers

Posted: Sun Aug 07, 2022 7:41 pm
by Daniel
Yeah, of course, just try it out, measure and listen. But I'm lazy and busy, so: has any of you ever experimented with restacking speakerformers?
With my AVCs I feel without experimenting with different stackings I wouldn't have gotten half of what they offer.
What does the math say in that case? With 800Hz and 7k55 crossover frequencies minimum inductance shouldn't be an issue?

Re: Restacking lams of speakerformers

Posted: Sun Aug 07, 2022 8:35 pm
by dave slagle
it simply comes down to inductance. if you have an excess then increasing the gap is worth a try.

dave

Re: Restacking lams of speakerformers

Posted: Tue Aug 09, 2022 7:33 pm
by Daniel
Is it just about changing the lf rolloff due to a change of inductance or also trading that for a linearized BH curve?

I restacked the laminations in the tweeter attenuator to 3x3xrest of themx3x3. As it took me around 24h between listening, starting to restack and finishing and listening again the subjective results are questionable:
No change in tonality. Sharper stereo imaging. Voices moved up by 50-70cm.

Re: Restacking lams of speakerformers

Posted: Thu Aug 11, 2022 6:06 pm
by Daniel
Changed stacking around a bit more and arrived at 2x2xrestx2x2. My narrow sweet spot got narrower than it was before with even less of a grey area.
Stero effect more defined, more details.
And that silver sound you get when changing from an AVC with copper to silver windings. That unique type of softness without losing details.

Re: Restacking lams of speakerformers

Posted: Sun Oct 02, 2022 5:59 pm
by Daniel
SlagleformerVergleich2.png
SlagleformerVergleich2.png (6.32 KiB) Viewed 10629 times
Top: autofomer as delivered 3x3 stacking, in circuit, mid driver replaced with 16R resistor.
Bottom: stacked 1x3x3xrestx3x3x1

Level is the same, this is how the software displays multiple graphs. Driving it from a 50R output impedance generator probably makes this all monkey business.

Objectively there is no difference but how do you measure "the background choir on these two records are way more prominent in the blue graph and before I couldn't tell it was three singers"?

Re: Restacking lams of speakerformers

Posted: Mon Oct 03, 2022 1:50 pm
by dave slagle
It seems odd that the change in the gap would effect the level in the measurements and it really needs to be driven from a ~1Ω source for reasonable results.

My guess on the sonic results have to do with the lower inductance value of the 3X3 creating a high pass for information beyond what the speaker can reproduce thus decreasing doppler distortion. I do think that my earlier thoughts about the 3X3 and other stackings changing the sound has more to do with the overall change in L value rather than any linearization to it. When it comes to any form of alternate stacking you introduce a frequency / level dependent non-linearity so various configurations can modify behavior but none of them are as linear as a discrete gap.

dave

Re: Restacking lams of speakerformers

Posted: Mon Oct 03, 2022 7:17 pm
by Daniel
Level was the same. When I choose the "display multiple graphs" option they are spaced apart and there was one with a measurment glitch in between that I deleted. I'll hook up a chipamp and measure again.

Re: Restacking lams of speakerformers

Posted: Sat Oct 08, 2022 8:54 am
by Daniel
Yesterday I was at a friends place who has a similar setup as mine. We restacked his tweeter speakerformers and didn't notice any change. :lol:

Re: Restacking lams of speakerformers

Posted: Sat Oct 08, 2022 12:31 pm
by dave slagle
interesting.

My two questions on the situations:

Was the shunt resistor the same in both cases?

What was the nature of the crossover driving each autoformer?

dave

Re: Restacking lams of speakerformers

Posted: Sun Oct 09, 2022 1:31 pm
by Daniel
My case: manganin film shunt resistor, 6dB serial crossover, 211 output tube through nano transformer

His case: manganin wire trimmer, dbx DSP crossover, safety cap before autoformer, 1875 chipamp

(His crossovers are still in the figuring things out phase, which is a lot easier with a DSP. The autoformer is in as a safety feature to not fry the Onken Tweeters. The trimmer makes it easier to set impedance to what ever we think is appropriate in the moment)