LC_Tuning-Outside_the_Box

I've never done this, but I saw a design idea in EDN to slap on a negative resistor circuit onto a LC and let it self oscillate. This would be instead of sweeping it.

Nope. You're generalizing... and you're wrong. This thread has been about seeking a way to tune an LC at a low enough frequency that component values are unwieldy. E1 has a |gain| < 1. Write the equations. They'll be edifying. ...Jim Thompson

Indeed! Part of my thought processing of this thread was to possibly simply turn this into a VCO... but I haven't gotten that far yet. ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

On a sunny day (Thu, 21 Nov 2013 12:33:10 -0800) it happened John Larkin wrote in :

Bull.

Floating? So you never heard of transformers either.

LOL use class D, now MORE inductors, NOW it is getting fun :-) Switch at a GHz, I like it,

Why not jut tel us what yo uare doing, tell the custardcomer to *!! , F35 will never fly anyways. Too bug free.

(Only the bugs fly).

Hi Tim, My tongue was a bit in my cheek.

d Ohh, I'd like one of those too! Say do you know of 'my' poor man's spectrum analyzer? Take your digital 'scope, trigger right up at the tippy top of the signal, set it in average mode with max number of averages. And then have it do t he FFT. Much cleaner than a single sweep, *BUT* the vertical scale drops o ff twice as fast. So if the 'scope says 5dB/div it's 'really' 10dB. I'm w aiting for the 'scope that can average multiple FFT's... (FFT first and the n average)

The only type of things I have that are 'unique' in any way would be optics type stuff. (And most of that centered around 780-800 nm.. Rubidium stuff .) There's diode lasers and driver's I didn't use, All sorts of partially 'failed' absorption cells, Rb discharge bulbs, lots of (cheap) optics piece s, (interference filters, polarizers, 1/4 lamda plates, some lenses) But n ot very useful unless you want to get into atomic spectroscopy.

And of course if I ever try the VCO from Y5V ceramic's I'd have to do it fo r myself... But I'm still interested in what someone else does/ finds.

George H.

Huh? What does "consumer grade" have to do with it?

Floating is NOT required, see link below

Again! What does "consumer grade" have to do with it?

I can do that, probably already have a perfect amplifier in my files back when, as a youngster, I rolled my own amplifiers from scratch.

For analysis, see...

...Jim Thompson

Unfortunately, my DSO is of the vintage where you physically need the board with a load of chips on it to do FFTs and stuff, not like today where it's a passkey dongle, or just built in from the get go.

Hmm... interesting... The sampled peak, and related harmonics (periodicity and sampling theorem at work) should be normal, and all other frequencies should be 1/N (or at worst, 1/sqrt(N) for uncorrelated noise) for N averages, no? I can't imagine averaging would be all that beneficial on a spread spectrum (FM) signal, for example, or in general, any signal with low autocorrelation over most any sample period (i.e. time between trigger events).

Heh neat, rather special purpose of course. Unless Phil or someone else here wants 'em. ;-)

I figured someone would be interested. The others I think are just stewing in their "rawr rawr we can take these measurements ourselves donations feh!!" mindset. Tsk, missed opportunities ;-)

Cheers,

Tim

Here's half of the full H-bridge amplifiers that I built for my 1977

I pulled out the stock AM/FM radio, replaced it with a nice Panasonic tuner and built full H's for each side, front and rear... probably why my hearing is so bad now... Rossini's "La gazza ladra" (my favorite of his overtures), and a lot Dixieland, at full power ;-)

Sorry for the image quality, an old C-size from 36 years ago.

Wonder if anyone can decipher the gimmicks in this amplifier? ...Jim Thompson

My TDS744A from 1995 can do that.

Well, the cheap Rigols are a bit disappointing FFT-wise, but OK if you don' t look at HF side (Keep FFT zoom above 2 and zero near DC). cheap TEK's ar e ok. I haven't gotten to play with better DSO's. The average function on a DSO is a must! A 'poor mans' lockin is even easier.

Yeah mostly I use it to look at amplifier response, So looking at noise, bu t then (as you say) correlated interference can also shows up.

Nah, I can only give you semi broken cells (some impurity gas in them, air, water?), that we can't sell. If Phil was interested he'd get good ones. All the optics I've got is cheap stuff, but making things work with cheap s tuff is half the fun!

George H.

I just ordered a "trial lens set" from ebay. It's about 80 lenses, pos/neg/cyl, and the eyeglass frame thingie, in a nice wooden box. I can now figure out my own eyeglass prescriptions and order glasses from Zenni. And then use all those lenses for electro-optical stuff.

About half the time that I have an optometrist write me a script, it's wrong. And different guys come up with very different numbers.

Den lørdag den 23. november 2013 03.27.36 UTC+1 skrev John Larkin:

t do

'm

and

on't look at HF side (Keep FFT zoom above 2 and zero near DC). cheap TEK's are ok. I haven't gotten to play with better DSO's. The average function on a DSO is a must! A 'poor mans' lockin is even easier.

ther

se)

l,

time

but then (as you say) correlated interference can also shows up.

eap)

some

se

ir, water?), that we can't sell.

p stuff is half the fun!

getting a pro to do it is free here if you buy glasses, if you just want the numbers its usually something like $20

you must be answering the better/worse questions differently

unless it is done with some electronic gizmo, which I've heard is pretty much only good as a rough guide

-Lasse

Certainly not on purpose.

They should give people a couple of knobs to turn, and let them find the best settings. Damned control freaks.

negative resistor circuit onto a LC and let it self oscillate. This would b e instead of sweeping it.

ply turn this into a VCO... but I haven't gotten that far yet.

If Jim hadn't kill-filed me he'd know I'd pointed out the same thing on the 17th November 2013 at 05:06:39 and 05:47:53 UTC+11, in the original "Tunin g an LC" thread.

My low-distortion oscillator uses this trick (which works in LTSpice simula tion, but not yet in real life, because I haven't get solered the bits toge ther). It needs a 90 degree phase shift network (which Jim doesn't seem to have noticed yet) and a high-output-impedance - current source - drive to p ush the artificial reactive current into the tank circuit where it looks li ke either a parallel inductance or a parallel capacitance, either positive or negative depending on your point of view and the sign of gain.

Our system is tightly regulated by the People who Know Better. For example, the minimum thickness of a glass lens is set to break your nose, by law. So, nobody gets glass.

It's pretty easy to measure your own eyes to 0.1 diopters with a ruler, and a printed test pattern (and possibly a set of reading glasses, if you're far sighted). Just adjust your focal length to within arm's length with external lenses (if it isn't already there naturally), then slide the pattern up and down the ruler like a trombone to find the focal length.

You hold the ruler against the bridge of your nose, and add in the distance from there to the surface of your eye.

If you're astigmatic, you'll have to rotate the test pattern for best angle, note the angle, and measure at that angle and at 90 degrees to it too.

The optometrists here round off to 0.25 diopters, so it's dirt simple to best their accuracy. It's not really a fair fight--you can actually SEE, and they have to guess.

I can measure myself with 0.05 diopter repeatability.

Now I just tell the optometrist what my Rx is. They never believe me, but I'm always right.

Cheers, James Arthur (posted from a Droid tablet, as an experiment)

your point of view and the sign of gain.

People have been implementing oscillator circuits using a 90 degree shift to obtain oscillator for decades, what makes your so special? I mean, I can do that with a jfet or a power mosfet with a basic rc phase shift circuit and obtain very clean sinusoidal waves.

Maybe you should repost your schematic since it has been a while since most of us have seen that thing of beauty, as you keep pointing out.

Jamie

While developing a full-duplex wireless hi-fi headset for the cell-phone industry; I used a CD player to highlight the link's sound transfer capability and preferred Dvorak's Four Seasons, especially the first few bars with its violins. To highlight the excellent quality of the link, I played the borrowed CD through it over and over and over ...until one day I couldn't find that CD ANYWHERE! Before the incident, I had heard rumblings from some of the lab personnel, but ignored them for the preference of listening.

By the way, the link cost pennies, could be made using discrete before moving to silicon, used a 'button' hearing aid battery, and could run 40 hours on it. passband was something like up beyond 15kHz with less than

Yep. I have all kinds of war stories where what I suggested was simple, efficient and cheap, but my suggestion was ignored because it lacked some flashy name, like 16-bits ;-) ...Jim Thompson

Figuring out the spherical correction is easy, and Zenni is so cheap that you can afford to bracket some pairs to see what's best. I need to figure out how to specify the cylinder correction for astigmatism. Apparently there are two conventions, positive and negative cylinder, so any prescription is inherently ambiguous. And prescriptions are full of incrementals and non-standard abbreviations.