<!DOCTYPE html>
<html><head>
<meta charset="UTF-8">
</head><body><p>One question I'd ask is what type of zeners were you using? Your typical run of the mill zener has pretty bad accuracy. There are precision ones that are better. Also there is something about lower voltage zeners not working as well as higher voltage ones if I remember correctly.</p><p><br></p><p>Jay S.</p><blockquote><p>On April 28, 2017 at 12:47 AM Tom Bugs <admin@bugbrand.co.uk> wrote:</p><p>One thing I wondered about Zeners a couple of years ago was whether<br>different types gave different results - I wasn't entirely sure whether<br>I was barking or looking at the wrong parameters though!</p><p>I *think* that BZX84 type were worse than BZV55 for example. One<br>parameter in the datasheet I'd wondered on was the capacitance measured<br>at 1MHz.<br>But I now can't remember if this changed the oscillating amplitude at<br>all freqs or just at the low freqs which is what I was trying to figure<br>(as per current discussion)</p><p>Barking up the wrong tree?</p><p>On 27/04/2017 18:25, Tom Wiltshire wrote:</p><blockquote><p>On 27 Apr 2017, at 16:03, Mattias Rickardsson <mr@analogue.org> wrote:</p><blockquote><p>On 27 April 2017 at 16:57, Richie Burnett <rburnett@richieburnett.co.uk> wrote:</p><blockquote><blockquote><p>But if the gains match well at 0dB, pretty well at -20dB, and not so well<br>by the time you're down at -80dB, then you'd expect resonance to drop off at<br>lower frequencies.<br>Or it could do the opposite, depending which of the two integrator gains is<br>lowest/highest! Like in that paper you referenced about varying SVF gain by<br>manipulating the integrator gains.<br>True!<br>This is intriguing. Does this indicate that the loss of resonance<br>towards the bottom end is actually caused by another effect than the<br>gain mismatch? Possibly capacitor leakage (e.g., through zeners) or<br>something else?</p></blockquote></blockquote></blockquote><p>There are three potential causes we've identified:</p><p>1) Phase lag<br>2) Gain mismatch being worse at greater attenuation<br>3) Zener current robbing the integrators.</p><p>The trouble is *all* of these effects operate in the same direction, making the filter less likely to resonate at lower frequencies.</p><p>How much of each is causing the effect we're seeing is still yet to be determined! It's quite possible that it's a mixture of all three and that sorting out any one might give an improvement but won't fix the problem on its own.</p><p>Any ideas for experiments we could try to find out which is the worst offender?</p><p>Tom</p><p>_______________________________________________<br>Synth-diy mailing list<br>Synth-diy@synth-diy.org<br><a href="http://synth-diy.org/mailman/listinfo/synth-diy">http://synth-diy.org/mailman/listinfo/synth-diy</a></p></blockquote><p>--<br>BugBrand LTD<br>UK company No. 07199808<br>VAT No. GB 988 2629 57<br>1 Ninetree Hill<br>BRISTOL BS1 3SB<br>United Kingdom<br>www.bugbrand.co.uk</p><p>_______________________________________________<br>Synth-diy mailing list<br>Synth-diy@synth-diy.org<br><a href="http://synth-diy.org/mailman/listinfo/synth-diy">http://synth-diy.org/mailman/listinfo/synth-diy</a></p></blockquote></body></html>