Path: ...!eternal-september.org!feeder2.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: john larkin <jl@glen--canyon.com>
Newsgroups: sci.electronics.design
Subject: Re: 80dB now but still needs improvement at 1KHz
Date: Tue, 19 Nov 2024 12:16:05 -0800
Organization: A noiseless patient Spider
Lines: 50
Message-ID: <ibqpjjlcta2qpdrqqthpl10e7oej20np9c@4ax.com>
References: <vg9pnu$ekhh$1@solani.org> <vga6jh$suru$1@dont-email.me> <vgatph$46a$1@nnrp.usenet.blueworldhosting.com> <vgbv7l$177js$3@dont-email.me> <vgekth$1anu$1@nnrp.usenet.blueworldhosting.com> <1r37pyc.1bodve2fz5t4wN%liz@poppyrecords.invalid.invalid> <vhiasv$54s$1@nnrp.usenet.blueworldhosting.com> <1r3a1m9.1lg1mngftnegwN%liz@poppyrecords.invalid.invalid> <vhife4$186c$1@nnrp.usenet.blueworldhosting.com> <vhik6t$1v4ua$1@dont-email.me> <o7npjjhbfgknlohffqpr99uv44r0imkb4o@4ax.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Injection-Date: Tue, 19 Nov 2024 21:13:42 +0100 (CET)
Injection-Info: dont-email.me; posting-host="02fde32b45fa01503e0808e8ceb9b45a";
	logging-data="2120719"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19UMEMi8LlEHXYv7YpwTRSH"
User-Agent: ForteAgent/8.00.32.1272
Cancel-Lock: sha1:dgMqm1VJtOMDNRn4HFqhgutp3Oo=
Bytes: 3540

On Tue, 19 Nov 2024 13:50:46 -0500, Joe Gwinn <joegwinn@comcast.net>
wrote:

>On Tue, 19 Nov 2024 19:12:18 +0100, Jeroen Belleman
><jeroen@nospam.please> wrote:
>
>>On 11/19/24 17:47, Edward Rawde wrote:
>>> "Liz Tuddenham" <liz@poppyrecords.invalid.invalid> wrote in message news:1r3a1m9.1lg1mngftnegwN%liz@poppyrecords.invalid.invalid...
>>[...]
>>>>
>>>> The startup transient of the FFT would have to be included, so that may
>>>> set a limit to the accuracy of the simulation measurements.  Even if you
>>>> start on a zero-axis crossing, you will get spurious results because of
>>>> the finite length of the sample.
>>> 
>>> Yes I understand that. It's actually a sinewave multiplied by a rectangular function.
>>> That's why I try to use a sample length of many tens of seconds.
>>> A window function might help but then I'd have to decide which window to use.
>>> 
>>>>
>>>> This can be demonstrated by doing FFT on samples of fewer and fewer
>>>> cycles from the middle of an apparently pure sinewave.  As the sample
>>>> length decreases, spurious harmonics begin to appear, even though they
>>>> don't exist in reality.
>>
>>An FFT implicitly connects the end of the recording back to the
>>beginning. It's sufficient that the number of periods is integer.
>
>If testing with sine waves that all fit with an integral number of
>cycles per window, the FFT noise floor will be determined by the
>precision of the arithmetic used. 
>
>But there is often a splice error at that join because the fit into
>the window is approximate, which is a good reason to use a window
>function.  The default choice in the radar world is Taylor.
>
>Joe Gwinn

I wouldn't trust LT Spice for simulating a low-distortion oscillator.

Run a sim and zoom up on the top of some sine wave. It looks all
segmented and ratty.

I was just simulating something and had a KHz range loop oscillation
that refused to be compensated. Turns out I had a BI behavioral
current source driving an inductor, and that oscillated all by itself.
A G current source does the same thing.

At really low distortion, the opamp models can getcha too.