[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [N8VEM-S100:2385] Z-80 V2 component questions
- To: email@example.com
- Subject: Re: [N8VEM-S100:2385] Z-80 V2 component questions
- From: David Riley <frave...@gmail.com>
- Date: Mon, 10 Feb 2014 14:17:21 -0500
- Authentication-results: gmr-mx.google.com; spf=pass (google.com: domain of frave...@gmail.com designates 2607:f8b0:400d:c00::236 as permitted sender) smtp.mail=frave...@gmail.com; dkim=pass head...@gmail.com; dmarc=pass (p=NONE dis=NONE) header.from=gmail.com
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=content-type:mime-version:subject:from:in-reply-to:date :content-transfer-encoding:message-id:references:to; bh=RtIKXkNPjrrJp8bG6/Orzq6+inJmE/kWqIC1417r8cU=; b=Cq05ROp68U7O+5huJRxOAWTPcwVo93DghTFUFuw7G7eg0GMBGyBELa8D2ddhdH979y 15g31aoNW4IdDDbTzG8f6YwzxBvwSYOM/dX5FuWE+LQqhEOgS1eydOZj/yP9AM1PHN77 y1N8CL2bXPPL4DTjwGmWkqW/rTLcUYGDrUSPZcWmcP3bPLt6KliwOAjP5PnoBn7YlZMZ TzBW6Xzndxvay10EDiagJFVr3rUTpCYk0G2P0zcqCgwjwMTG7gT453hIdXipOim7OYeE +zDJJ428KFE4suaUa8vUlwuD1DiIS3MTRVOV1YPqtME6lWoRI2nrO0oyMM+asnJ6+Ukx 2jvg==
- In-reply-to: <BAY175-W4439918335B34198833F33B8900@phx.gbl>
- References: <firstname.lastname@example.org>,<email@example.com> <BAY175-W4439918335B34198833F33B8900@phx.gbl>
On Feb 10, 2014, at 1:58 PM, Crusty OMO <crus...@hotmail.com> wrote:
> That advice of using a decoupling capacitor .1uf near the regulator output pin is important. I have seen regulators go into oscillation without it.
Just don't take it as gospel for all regulators; some regulators go into oscillation WITH a low-ESR cap (like a ceramic) on the output, requiring aluminum or tantalum electrolytics instead. Best to read the regulator's datasheet and see what they recommend.
A SMALL low-ESR cap (tantalum, polymer or ceramic) is often used to decouple the larger electrolytics at the supply output, but some modern regulators are good and fast enough that they don't need more than 22uF or so on the output, and it's tempting to just do it with a large ceramic; that can be bad news if the regulator isn't designed for that, because a regulator control loop that's designed assuming a high-ESR output (which essentially acts as a high-pass filter for the feedback) can easily oscillate at that point.
> Big electrolytic capacitors don't work well at high frequencies, so a decoupling capacitor must be used next to them.
> The big caps are great storage caps, but very sloppy at stablizing the voltage on the rails.
This is exactly right. It's all a matter of effective serial resistance (ESR); aluminum electrolytics usually have terrible ESR (though they're getting better), and tantalums somewhat less so, but tantalums are finicky and somewhat expensive (and have an awful tendency to blow up when they get old).
Ceramics have vanishingly low ESR, but they are small, so you usually have to balance your needs. Polymer caps are close to ceramics and are a bit more physically durable, but they're also more expensive. Polymer is generally better for analog paths (especially audio) because ceramics have a tendency to be noisily microphonic (every capacitor is microphonic, but ceramics especially so, partly due to the piezoelectric effect).
Inductance plays a part as well, which is why it's necessary to have small ceramics (or polymer caps, which are almost as good) distributed around near the parts you need to decouple. Otherwise, the power/ground planes and supply lines contribute enough inductance to not only counterbalance the capacitance, but also resonate with it, which causes ringing. One cap per chip is overkill, but it won't hurt (caps are cheap, though you'll spend more time soldering them in).