coil impedance

If you are simulating something where the signal-to-noise ratio is the primary concern, I would have thought it was important to take the actual noise into consideration.

"Some design engineers may not realize that VCR even exists"

Now who could that be?

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

LT Spice includes noise in a frequency-domain analysys, but not in time domain.

That's OK for linear systems.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Sort of interesting but not all that specific. Thin metal film resistors are better than thick film resistors,but we knew that anyway.

Probably you. You do seem to have become a believer very recently.

But you haven't been able to find a write up that quantifies it. And oxide film tunneling only happens if the if metal-to-metal contact hasn't been tightened up hard enough to drive the high points through the oxide film. That's why SMA connectors can be better than SMB or SMC.

Johnson noise is broadband. The harmonic of interest in evaluating a low distortion sine wave oscillator are rather narrow band.

You are comparing apples and pears as your "5 Kc.s bandwidth" makes clear. if you have been cribbing from a more modern source it would have been a 5kHz bandwidth.

Such people tend to post sims with inductors that don't have any parallel capacitance, as John Larkin has been known to do. They tend to get snooty when the sim contains by-pass capacitors.

More qualitative pontification.

At DC you have to worry about 1/f noise, which starts ramping up below about 10Hz with bipolar juntions, about 100Hz with FETs and kHz with MOSFETs (though that is very process dependent).

A 1kHz reference sine wave is in a different ball-park.

When you get down to signals as low as -150 dBm there will be Johnson noise and intermodulation products contributed by the components within the oscillator. Selecting individual harmonics with a narrow-band filter will give an unrealistic number that doesn't represent the total unwanted output.

You have no idea what I did, so stop posting offensive nonsense.

I have written a calculator based on a spreadsheet which gives the RMS noise as a voltage and as dBm from user inputs of resistance, bandwidth and temperature. I chose 5 Kc/s as a tolerably low figure for audio work; it would be a suitable measurement bandwidth for a 1 Kc/s signal if you wanted to include the third harmonic. I chose 50 ohms as it is the lowest output impedance in common use - for audio work 600 ohms might have been more appropriate but the figures would have been even worse.

I have always written "Kc/s" and "Mc/s" because I regard them as more intuitive than having to remember who the units have been named after. I presume you didn't notice this in all my previous posts.

If someone wants to claim distortion products at or below the Johnson noise, they need to justify those claims. Selecting a narrow-enough bandwidth for the measurement could give any figure you like.

I agree - but there was no indication that we were talking about signals in the kilovolt region. I took the practical example of +20 dBm as being the highest signal level normally encountered, which is 150dB above the lowest Johnson noise normally encountered.

With exceptonal signal voltages and source resistances the 150 dB figure can be bettered - but anyone making claims in that region needs to be explicit about how the measurements were made, otherwise we could be misled into assuming they were using the 'normal' criteria for such measurements.

I'm pretty confident that you don't have much of an idea what you did either. That's probably offensive, but sadly it isn't nonsense.

But if you are looking for distortion in a sine wave, what you need to know is the harmonic content generated by the distortions.

The harmonics apear at very specific frequencies, and and if the oscillator generating the sine wave is operating at a stable frequency, these harmonic will appear at very specific multiples of the base frequency

If you frequency meter says the sine wave is at 1kHz, and keeps on saying that for more than a second at a time, the bandwidth of the sine wave is going to be narrower than 1Hz, not 5kHz, and the fifth harmonic is is going to less than 5Hz wide, not 5kHz wide even if it is at 5KHz.

I don't have any trouble translating them into the currently accepted units, and hadn't actually noticed.

I've got a science degree, and even back in the 1960's we got indoctrinated about using the standard units, and keeping up with the internationally agreed standard units. "Cycles" is no less arbitrary than "Herz". For an English speaker it may be more intuitive, but for me a cycle has two wheels - my wife gave me a unicycle once, and I managed to learn how to ride it, but it didn't reshape my vocabulary.

And selecting 5kHz as the bandwidth you like is giving you an amplitude that fits your argument. It's nonsense, but superficially plausible nonsense. For a very generous level of superficiality.

I build stuff, and often sell it. Sometimes I Spice something that has no immediate use, but usually there is a real product in mind.

How much of your stuff gets built or sold?

There is occasionally justification for putting parasitic capacitance across inductors in sims, but usually not. Engineering instincts can ballpark order of magnitudes enough to know when it's not worth considering.

I don't put bypass caps across voltage sources in Spice. Do you?

On real multilayer PCBs, I don't think I've ever seen a power pour that had too little bypassing. There's usually too many caps by at least 2:1. Well, my favorite bypass cap costs one cent.

This weekend I'm Spicing a Rogowski coil current pulse digitizer. Proposal is due next week. I ordered a coil from Amazon to play with. Amazon is awesome.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Stop making sense! Just stop it.

I never liked Hz. There's nothing wrong with megacycles.

Remember uuF?

That -150 dB oscillator sim is just using Spice as a video game, to waste time for people without much to do.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

It's been a consideration in my high voltage supplies and AC power meters for decades. Resistors in series make sense for many reasons.

I tested a bunch of high-meg surface-mount cermets for VCR and excess/shot noise. That was hard, specifically the noise part.

You work hard to generate insults, but they are weak and childish.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Liz doesn't have much of an idea what she is doing, and she's offended when this gets pointed out

It is surprising what makes sense to John Larkin. Not as surprising as the fossil carbon industry counter-propaganda about climate change which he can swallow without gagging, but surprising enough.

Picofarads are so counter-intuitive

John Larkin comforts himself with all kinds of superficially plausible delusions.

Jim Williams played that game without even having a Spice console to play on.

None in recent years. When I was in paid work, we stopped working on products if we didn't expect to be able to sell them, but I did quite a lot of work on stuff that was already selling, but would sell faster if we could make it better.

Spice is supposed to train your instincts. It is a silly claim. Instincts are inherited. What you learn by doing isn't any kind of instinct, but if you don't think about what you doing you can still learn stuff at an unconscious level.

Since I tend to put ferrite beads in series with each of the by-pass capacitors when power supply rejection needs to be good, I have been known to do it. Spice can pick up unexpected resonances.

Everybody plays safe. I've seen an ostensibly well-by-passed hand-wired board upset one of my junior engineers - he'd bypassed one of the op amp power pins to the wrong place on the power track. When I got him to move the capacitor to the right place on the track the problem went away.

Putting a quite small inductance in series with the bypass capacitor isn't always a good move.

I've sent one printed circuit layout back to the draftsman have a similar potential problem fixed. Both events were back in the late 1970's.

And buying your coil from Amazon gives you complete control over how big the parallel capacitance is?

Am 15.11.25 um 17:03 schrieb john larkin:

Cermet is usually a pseudonym for thick film. That can be easily much worse than carbon composition, esp. for 1/f noise, preferably when there is DC across it. Thin film rulez. I use Susumu SMDs as default (DK).

Gerhard

...

I can remember one of the presenters on

There is clearly more than one person here who thinks that this would be a better group if you could leave completely Bill.

I've worked in many places where you wouldn't get a job. Even if you could do the work. And if you did get the job you'd probably end up leaving due to you thinking you'd been badly treated, when in fact it was you who had treated others badly.