Identifying crossover frequency from schematics.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

Zeccar

Member
Joined
Jan 7, 2022
Messages
6
Location
Paris
Bonjour !
First of all I am very new to speaker design. I am trying to get proper speaker replacements for a pair of scavenged Sony SS-A1L (2 way speaker, LS3/5A clone). Data is scarce but I found some details on sensitivity, impedance etc. of the original tweeter. I also found the service manual with the Xover schematics. I am trying to find the crossover frequency but tbh I can't even tell its type/order. Could you folks help me out ?
Thnx :))

 

moamps

Well-known member
Joined
Sep 24, 2005
Messages
2,290
Location
Croatia, HR
The woofer has a crossover 6dB/oct, tweeter 18dB/oct. The real crossover frequency is somewhere between 1 and 3 kHz, IMO, but it can only be said exactly when the speaker is measured with a measuring microphone. Calculation/simulations in such cases can be inaccurate because the driver impedances and sound pressure depend on the frequency (as well as other parameters). I hope it helps.

1641683526374.png
 

Zeccar

Member
Joined
Jan 7, 2022
Messages
6
Location
Paris
Thanks a lot ! That's exactly what I was looking for. I guess according to your data the safest bet would be to go for a tweeter with a low Free air resonance frequency, one that would carefully sit underneath the 1k region. The Seas 27/TBC-G happens to have the same sensitivity and dome size as the original Sony one, so that pretty much settles the issue.
I guess it's more complicated to find a proper woofer replacement since most available units seem to have an impedance rating of 8 ohm. Installing one would shift the crossover to a higher frequency, but that could yield interesting results in boosting the midrange, ofc the graph you provided will help a lot in anticipating as much as possible the way the crossover would handle a given frequency response. At any rate, I'll keep you updated. Thanks again !
 

Zeccar

Member
Joined
Jan 7, 2022
Messages
6
Location
Paris
Update : I installed a Seas 27TDFC tweeter in one of the speakers this weekend.
+ Clarity has been dramatically improved. Nice sparkle and tiny details here and there.
+ Voices are a kinda harsh though. They feel overly detailed. esp from recordings that already have overemphasized treble. worst case scenario : Supertramp – Dreamer.
+ Midbass and depth of male voices (and overall midrange) is recessed. While it's not distorted, the sound is dry and pretty treble forward. Not really up to my expectations.

It almost feels like the tweeter stole all the juice from the woofer. Safe to say that the experiment was a failure. 🥴 Will prob. sell the parts.

* Also : there is an inconsistency in the Sony manual : after checking the real crossover circuit : the resistor R4 is 4.7 not 47 ohm as shown on the schematics.
 

abbey road d enfer

Well-known member
Staff member
GDIY Supporter
Moderator
Joined
Jan 22, 2008
Messages
14,766
Location
Marcelland
Update : I installed a Seas 27TDFC tweeter in one of the speakers this weekend.
+ Clarity has been dramatically improved. Nice sparkle and tiny details here and there.
+ Voices are a kinda harsh though. They feel overly detailed. esp from recordings that already have overemphasized treble. worst case scenario : Supertramp – Dreamer.
+ Midbass and depth of male voices (and overall midrange) is recessed. While it's not distorted, the sound is dry and pretty treble forward. Not really up to my expectations.

It almost feels like the tweeter stole all the juice from the woofer
All this probably because the new tweeter has a higher sensitivity (efficiency) than the original.
The attenuator constituted of the 2.7 ohm and the improperly marked 47 ohm is probably to be modified. I would start by putting a 10 ohm resistor across the tweeter, and adjust to taste.
Anyway that would be a BandAid fix since proper x-over adjustment can only be done with proper test equipment, whatever subjectivist designers claim.
Replacement of the woofer is even dicier since the interaction between the box and the driver is quite complex, and improper matching can result in disastrous performance.
Contrary to what some unscrupulous vendors claim, putting a good driver into a good box does not always result in a good speaker.
 

Zeccar

Member
Joined
Jan 7, 2022
Messages
6
Location
Paris
All this probably because the new tweeter has a higher sensitivity (efficiency) than the original.
The attenuator constituted of the 2.7 ohm and the improperly marked 47 ohm is probably to be modified. I would start by putting a 10 ohm resistor across the tweeter, and adjust to taste.
Anyway that would be a BandAid fix since proper x-over adjustment can only be done with proper test equipment, whatever subjectivist designers claim.
Replacement of the woofer is even dicier since the interaction between the box and the driver is quite complex, and improper matching can result in disastrous performance.
Contrary to what some unscrupulous vendors claim, putting a good driver into a good box does not always result in a good speaker.
Thanks for the advice🙏 Tbh the sound is not that bad so a BandAid fix might very well do the trick. (simply put : less treble and more midrange presence and its pretty much done).
The woofer (130mm) installed by the previous owner works surprisingly well with the small sealed cabinet, given it's a cheap unit. Bass is tight, timbre is nice. Too bad the only graphs available are for the 8 ohm version.

I'll try the resistor tweak for the tweeter. I suppose that would shift the crossover point a bit higher ? I could then change the low pass coil (L1) to a lower value to account for that, and get more midrange coverage. Although the hole in the 1.5K region in the graph doesn't look promising… Hopefully the woofer will somehow handle it, but hey, I'm pretty sure it's not that simple… right ? 🤪

1643023726338.png 1643022389380.png
 

Zeccar

Member
Joined
Jan 7, 2022
Messages
6
Location
Paris
Update : Ok no more bandaid shenanigans. I managed to get BoxSim to work on macOS and came up with a new angle.
Ofc There's no trusting in a simulation, but a measurement setup is out of the scope for now. At the very least it's better than randomly replacing a component and praying for the best.

Current FR :
1643836035405.png


So basically I am going for a new crossover design. Not really aiming for a flat response, but for the LS3/5A one. Psychoacoustics over objectivity ? Well it makes sense not to rely on a 5" woofer to produce natural deep bass on par with the rest of the spectrum, also since it's for my home cinema setup, vocals are a definitely a priority. So I surveyed the original frequency response, also that of the Yamaha NS-10 (known for amazing vocals). At any rate, the FR is but one of many parameters, and this project is more of homage than a clone : bringing the copy back home. I'm actually looking forward to the original Sony cabinet and new drivers' very own idiosyncrasies.

Tweeter : Seas 27TDFC
New woofer : Monacor SPH-135 AD (used for the 60th anniversary edition of the Rogers LS3)

As you may know, there are many iterations of the LS3 : Here's the FR of the Falcon LS3/5A gold badge.
1643836709051.png
And the current version of my simulation, with a less pronounced mid-bass bump. The ugly 250Hz bump comes with the driver.
1643836923753.png

It's not a 100% match for now. (white is mine, black is the Falcon one).
1643837770830.png



A very simple crossover indeed… Values are approx. I have no idea if that would work or melt. Still on it.

1643837502059.png

Dunno about the phase, but the acoustic power looks really bad.

1643837105456.png

1643837200699.png 1643837244186.png 1643837290934.png
1643837315224.png 1643837339721.png
 

Disco Volante

Well-known member
Joined
Jun 20, 2021
Messages
90
Location
Sweden
Interesting project. The predicted SPL-curve isn't exactly hi-end, but I guess that's not really the goal here.
The white curve looks significantly more listenable than the original Falcon curve, so you might be on to something;-)
I would increase the 2R5 in series with the tweeter (3R9?, 4R7?) to get less aggressive treble.
Also, the major bump and phase shift around 1200 Hz, just before the point of crossover, could probably be alleviated somewhat by coating the inner half of the rubber surround with whatever rubbery substance you have at hand (e.g. contact glue or shoe-repair or latex)
Happy tinkering!
 
Top