High Frequency Section of a A Type Crossover (Heritage)

[reinderspeter]

Can someone explain to me how the High Frequency Section of a A Type Crossover is supposed to work?

Normally you'd expect each section to connected to the + and - input but here the 2uF tweeter cap is connected to a point after the13uF Midrange cap.

I have never seen a schematic like this before and I would love to know the reasoning behind it.

[Don Richard]

Configuring a crossover in that manner has several positive benefits, primarily in keeping lower freqs away from the tweeter. The slope is 6 dB per octave for the first two octaves transitioning to 12 dB per octave below that.

For further explanation and graphs see Sound System Engineering by Davis and Patronis, 3rd edition. I believe they call it an "improved three way crossover".

[WMcD]

Of course I agree with Don.

You might want to check out PWK's article on the use of autotransformers. http://forums.klipsch.com/forums/p/59538/582694.aspx#582694

If I read correctly, it is mostly concerned with the midrange horn and a situation where 6 dB of attenuation is needed.

You will see that the pre-A type crossover design shown there does indeed connect the tweeter through one cap, directly to the input. Perhaps what you had expected.

We can only guess about PWK's thinking in moving that tweeter connection from the input from the amp to the output of the 13 uF cap. It should have no effect on the sound output of the tweeter.

But as Don points out. Now, by the alteration of wiring, the input to the 2 uF is already being attenuated below 400 Hz. Maybe that is not needed -- save for gross overloads or failure of the 2 uF cap. However, no matter how small the benefits might be, they are free of charge with no extra components being added.

It's free. So why not?

Wm McD

[Bacek]

Isn't this a third order filter? Series capacitor (13uf), coil to ground (autoformer) and then next series capacitor.

[djk]

The inductance is too high to figure in as a crossover element.

[greg928gts]

I had a customer write to me the other day, regarding a DHA crossover, saying that he changed the location of the tweeter cap to be wired AFTER the midrange cap and he thought it sounded better.

Greg

[reinderspeter]

Thanks for all the answers.

Isn't this a third order filter? Series capacitor (13uf), coil to
ground (autoformer) and then next series capacitor.

This is what I was thinking too.

Either that or it is 13uF and 2uF in series resulting in a 1.75 uF cap and consequently a higher cut of frequency.

The inductance is too high to figure in as a crossover element.

Does anybody know the inductance of the autoformer?

[Don Richard]

Isn't this a third order filter? Series capacitor (13uf), coil to ground (autoformer) and then next series capacitor.

This is what I was thinking too.

Either that or it is 13uF and 2uF in series resulting in a 1.75 uF cap and consequently a higher cut of frequency.

The inductance and capacitance are too large to cause a 3rd order rolloff on the HF section, but sufficient to steepen the slope 2 octaves below the cutoff frequency. The HF capacitor value was adjusted during design to account for the circuit change.

[djk]

"Does anybody know the inductance of the autoformer? "

Most T2A measure around 40mH, it depends on how the core is stacked.

[Deang]

T2A Data Sheet attached.

I've tried to measure them using an inducatance meter -- the measurement is constantly on the move. I've given up trying to understand that one.

According to the data sheet, inductance varies depending on what tap you're on. Some of the values certainly seem low enough where the inductance would have an impact. Regardless, using an autoformer impacts phase -- something most don't think about.

As for the 2uF cap, moving it to input definitely changes the sound. I could never could figure out exactly what it was I was hearing. Sometimes it seemed like slightly elevated output, sometimes it seemed like it was something else. In the end I decided I liked it wired PK's way. "Different" is not "better" -- though it's often interpreted that way.

[ngk]

Am I correct in saying that positioning the tweeter capacitor after the midrange capacitor - at its output - maintains the constant impedance characteristics of the dhaxover and alk universal crossovers. If the tweeter cap (or tweeter circuit in the alk) is connected before the midrange cap, the impedance will fluctuate by a certain amount in these crossovers.

Nick

[djk]

If it's about 21mH from 4-0, trhen it's at least twice than (40mH) from 6-0.