experiences with ESL directivity?

[kavermei]

Hi Alex,

the AES article's setup is comparable to the ladder in Fig. 2. The author uses RCR T-sections, so he has a resistor in front of the middle section too. But I don't think it'll make much difference.

Kenneth

[bolserst]

Originally Posted by alexberg

In Frank Verwaal papers there is a table (Part #2 p.80) with the RC filters to be used for linear frequency response. Next figures show the implementation of the idea.
So, am I right, assuming:
1) for different stip widths the table and individual resistors per stip (Fig.#1) is applicable.
2) if the ladder structure is employed (Fig.#2) then all the resistors are equal, strip width are all the same and bolsert's spreadheet is applicable, as well as the AES article we discussed.

Hello Alexberg,

If you use individual resistors to feed each strip (Fig. #1) you will need increasing resistor values for each section to obtain a flat response. The resistor values shown in Frank's table are specific to his design. ie number and size of each segment, capacitance of each segment, target bandwidth for flat response.

As kavermei already mentioned, the ladder network (Fig. #2) with equal sectional widths and resistors is similar to the symmetric layout analyzed in the AES paper. All that is missing is the initial feed resistor.

The spreadsheet gives values for the asymmetric layout which the AES paper uses when deriving most of the equations. I modified your (Fig. #2) to show what feed layout should be used for the section sizes and feed resistances(R) calculated in the spreadsheet. I now realize I should have included it in the spreadsheet documentation since not everybody who uses the spreadsheet will have a copy of the AES reference paper.

Note that the resistor feeding the first section is half the value of the rest of the resistors in the ladder network. In practice, this resistor can be put on the primary side of the transformer where it can also help with saturation control. Just take the value (R/2) and divide by the square of the step-up ratio. Be advised that you will probably have to tweak the value of this first resistor to compensate for other factors not considered in the HF response as mentioned on the directions tab of the spreadsheet.

3) Transformer leakage inductance behavior at HF is ignored
4) Section to section capacitive coupling is ignored
5) HF loss due to the inertia of the diaphragm and/or the inertia of air in the holes in the stators is ignored.


If anybody is interested, I can post the ratios or factors to apply to the spreadsheet values for feed resistance and sectional capacitance if you want to use a symmetric layout. Let me know. Or, perhaps, a version 2 of the spreadsheet with separate tabs for symmetric and asymmetric configurations.

[Few]

If anybody is interested, I can post the ratios or factors to apply to the spreadsheet values for feed resistance and sectional capacitance if you want to use a symmetric layout. Let me know.

I'm planning on a symmetric layout so a spreadsheet version for that would be great! Thanks for the generous offer.

Few

[chinsettawong]

Hi Bolserst,

Thanks for sharing the spreadsheet.

Wachara C.

[kavermei]

Hi,

Originally Posted by bolserst

[...]In practice, this resistor can be put on the primary side of the transformer where it can also help with saturation control. Just take the value (R/2) and divide by the square of the step-up ratio.[...]

I would argue that it should be kept in the secondary circuit, since in the primary it will cause distortion.

I think it's better to control saturation by using a line-level highpass filter.

Kenneth

[alexberg]

Thanks a lot!
That's exactly what I thought, but it does not hurt to ask.
With the best regards,
Alex

[owenhamburg]

Having now owned Quad ESL 57's and Quad ESL 63's I have to say quads combination of filter and delay line in the ESL 63 is very impressive, for improving the treble dispersion. Unless you play with a pair of ESL 63's for a while you would not realize how good they are at treble dispersion because the bass is still very directive.

[bolserst]

Originally Posted by kavermei

Hi,

I would argue that it should be kept in the secondary circuit, since in the primary it will cause distortion.

I think it's better to control saturation by using a line-level highpass filter.

Kenneth

I agree, from a distortion standpoint you want to minimize the resistance in the primary circuit. For practical reasons, I usually keep a small primary series resistance to protect the amplifier just in case something unexpected happens or the filter network is accidently bypassed. Any additional resistance needed is put in the secondary circuit. I found that as long as the primary series resistance is < 2ohm the distortion increase is pretty minimal for typical M6 core material.

The reason I mentioned it was to point out that even if the first feed resistor doesn't seem to be in the circuit(on the secondary side connected to the first section) it might have been moved to the primary side where its function is not as obvious.

For those unfamiliar with the source of distortion kavermei is talking about, this post explains it a bit more.
Step-up transformer design

Also, for a visual aid, geraldfryjr had posted some plots showing the input voltage and distorted current waveforms for the primary circuit.
Step-up transformer design

[bolserst]

Originally Posted by Few

I'm planning on a symmetric layout so a spreadsheet version for that would be great!

I should have time to post an updated spreadsheet in a few days which includes the symmetric layout.

[kavermei]

Yes, I agree, 2 ohm should be okay.

Thanks for the spreadsheet, it's very handy!

Kenneth