Why would anyone post a response _before_ the bit they want to respond to? Please do try to make some sense Bill. My response was exactly where is should have been in the context of what I was responding to.
You still haven't pointed out the post where I "snipped out the bulk of the comment without marking the snip". Is that because no such posts exists? Either point it out or shut up.
A reference low distortion sine wave generator isn't going to be used in hostile environments.
Nobody makes much fuss about the frequency stability of these low distortion sine wave generators - I have worked out a scheme to add in a small controllable quadrature component to let you stabilise the frequency as well as the amplitude but it never generated any interest. The phase change in the Teflon film probably wouldn't be a problem in this application. It's not as if it is going to generate harmonics.
[Doesn't matter]
Will you stop the pointless bickering yet?
Here's a paper with an actual measurement of coil impedance over frequency:
"A Mixed Surface Volume Integral Formulation for the Modeling of High Frequency Coreless Inductors" June 2015 IEEE Transactions on Magnetics 52(3)
I'd have done some myself if I'd still had routine access to a VNA.
Jeroen Belleman
You need to respond to the entire message, not the message you've managed to create by truncating what was posted to create the message that you'd prefer to respond to.
I don't think that there's any point in producing the kind of response you'd have the wit to understand.
<snip>
Nothing was truncated except by yourself. If you believe I snipped (removed) anything from any post then you need to reference that post and the specific part of it or shut up.
If you believe that I ever mentioned tantalum bead capacitors anywhere else then you need to reference that post and the specific part of it or shut up.
Tasteless, but he did say "they let you."
Sexual aggression is a common trait in politicians; it's part of the power pattern.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Semiconductors have short thermal time constants. Opamps have internal nodes with tens-of-millisecond time constants and nonlinear junctions everywhere.
Most Spice opamp models are behavioral, nothing like realistic.
Resistors have thermal time constants and especially have voltage coefficients; the very best are a few PPM/volt and most are far worse.
Caps are nonlinear for lots of reasons.
But you'll never build your oscillator and would have no way to test it for sub-PPM distortion.
How do you propose to acquire the waveform to Fourier analyze?
Or make a filter with -150 dB distortion?
All nonsense.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
"john larkin" snipped-for-privacy@glen--canyon.com wrote in message news: snipped-for-privacy@4ax.com...
If I wanted to build a very low distortion 1k Hz oscillator then I'd build the circuit below. I'd be happy to see 90dB in reality if that could be measured. If better than 90dB can be measured then that's fine but the expense of doing so will likely start to increase exponentially.
Bill will be along soon to tell us who designed what and how he did this or that but we all know that anyway.
I don't claim any credit for this circuit except for the Q1 Q2 circuit which I arrived at by doing online research into how it might be implemented. AS3944 might be a good choice for that circuit. (This won't stop Bill, for the nth time, pointing out how further refinements were added but who cares.)
I wanted to see whether it could be done without thermistors, opto devices, lamps, FETs or similar devices. And also without large capacitors. After all if you wanted to implement it all on a chip then you don't want those kinds of components in the circuit if you can avoid them. And none of the FET based circuits I ever saw could match the LTSpice performance of this circuit.
Version 4.1 SHEET 1 3020 2548 WIRE -1936 -288 -1936 -336 WIRE -240 -208 -608 -208 WIRE 480 -208 -160 -208 WIRE -1936 -160 -1936 -208 WIRE -1840 -160 -1936 -160 WIRE -1648 -160 -1760 -160 WIRE 80 -160 -1584 -160 WIRE -608 -96 -608 -208 WIRE -240 -96 -608 -96 WIRE 80 -96 80 -160 WIRE 80 -96 -160 -96 WIRE -1936 -64 -1936 -160 WIRE -1936 -64 -2272 -64 WIRE -2400 -32 -2400 -288 WIRE -2272 0 -2272 -64 WIRE -2208 0 -2272 0 WIRE -2112 0 -2144 0 WIRE -608 32 -608 -96 WIRE -496 32 -608 32 WIRE -352 32 -416 32 WIRE -272 32 -352 32 WIRE -112 32 -192 32 WIRE -64 32 -112 32 WIRE 80 32 80 -96 WIRE 80 32 0 32 WIRE 144 32 80 32 WIRE 288 32 224 32 WIRE 336 32 288 32 WIRE 480 32 480 -208 WIRE 480 32 400 32 WIRE -1328 48 -1328 0 WIRE -1008 48 -1072 48 WIRE -880 48 -944 48 WIRE -1936 64 -1936 -64 WIRE -2272 96 -2272 0 WIRE -2224 96 -2272 96 WIRE -2112 112 -2112 0 WIRE -2112 112 -2160 112 WIRE -2000 112 -2112 112 WIRE -2400 128 -2400 48 WIRE -2224 128 -2400 128 WIRE -608 144 -608 32 WIRE -496 144 -608 144 WIRE -112 144 -112 32 WIRE -64 144 -112 144 WIRE 288 144 288 32 WIRE 336 144 288 144 WIRE -2400 160 -2400 128 WIRE -2400 160 -2512 160 WIRE -1072 160 -1072 48 WIRE -1008 160 -1072 160 WIRE -880 160 -880 48 WIRE -880 160 -928 160 WIRE -352 160 -352 32 WIRE -352 160 -432 160 WIRE 80 160 80 32 WIRE 80 160 0 160 WIRE 480 160 480 32 WIRE 480 160 400 160 WIRE 512 160 480 160 WIRE -496 176 -608 176 WIRE -64 176 -96 176 WIRE 336 176 304 176 WIRE -2512 192 -2512 160 WIRE -2400 192 -2400 160 WIRE -96 208 -96 176 WIRE 304 208 304 176 WIRE -1680 224 -1680 0 WIRE -1472 224 -1472 0 WIRE -1936 272 -1936 160 WIRE -1744 272 -1936 272 WIRE -1536 272 -1744 272 WIRE -1328 272 -1328 128 WIRE -1168 272 -1328 272 WIRE -1072 272 -1072 160 WIRE -1072 272 -1104 272 WIRE -992 272 -1072 272 WIRE -2512 288 -2512 256 WIRE -880 288 -880 160 WIRE -880 288 -928 288 WIRE -832 288 -880 288 WIRE -736 288 -768 288 WIRE -608 288 -608 176 WIRE -608 288 -656 288 WIRE -528 288 -608 288 WIRE -112 288 -448 288 WIRE 80 288 80 160 WIRE 80 288 -112 288 WIRE -1936 304 -1936 272 WIRE -1328 304 -1328 272 WIRE -992 304 -1072 304 WIRE -2400 336 -2400 272 WIRE -1072 336 -1072 304 WIRE 80 336 80 288 WIRE -1680 352 -1680 320 WIRE -1680 352 -1872 352 WIRE -1472 352 -1472 320 WIRE -1392 352 -1472 352 WIRE 80 448 80 416 WIRE 224 448 80 448 WIRE 48 464 16 464 WIRE -1680 480 -1680 352 WIRE -1472 480 -1472 352 WIRE 48 496 48 464 WIRE 80 496 80 448 WIRE 224 496 224 448 WIRE -1936 512 -1936 400 WIRE -1328 512 -1328 400 WIRE -1472 592 -1472 560 WIRE 64 608 64 560 WIRE 224 608 224 576 WIRE 224 608 64 608 WIRE -352 656 -352 160 WIRE -112 656 -112 288 WIRE 64 656 64 608 WIRE 480 656 480 160 WIRE -352 752 -352 720 WIRE -112 752 -112 720 WIRE 64 752 64 720 WIRE 480 752 480 720 WIRE -1936 800 -1936 768 WIRE -1936 800 -1984 800 WIRE -1984 816 -1984 800 WIRE -1936 832 -1936 800 WIRE -864 864 -960 864 WIRE -704 864 -800 864 WIRE -352 880 -352 832 WIRE -112 880 -112 832 WIRE -112 880 -352 880 WIRE 64 880 64 832 WIRE 64 880 -112 880 WIRE 144 880 64 880 WIRE 480 880 480 832 WIRE 480 880 144 880 WIRE -352 928 -352 880 WIRE 144 928 144 880 WIRE -960 944 -960 864 WIRE -912 944 -960 944 WIRE -800 944 -848 944 WIRE -704 944 -704 864 WIRE -704 944 -720 944 WIRE -704 1056 -704 944 WIRE -704 1056 -784 1056 WIRE -352 1056 -352 1008 WIRE -352 1056 -704 1056 WIRE -272 1056 -352 1056 WIRE -144 1056 -192 1056 WIRE 0 1056 -80 1056 WIRE -1680 1072 -1680 560 WIRE -1168 1072 -1680 1072 WIRE -960 1072 -960 944 WIRE -960 1072 -1104 1072 WIRE -848 1072 -960 1072 WIRE -752 1088 -784 1088 WIRE -1680 1104 -1680 1072 WIRE -752 1200 -752 1088 WIRE -1680 1232 -1680 1184 FLAG -96 208 0 FLAG 304 208 0 FLAG 512 160 vout FLAG -816 1040 vcc FLAG -816 1104 vee FLAG -1984 816 0 FLAG -1936 688 vcc FLAG -1936 912 vee FLAG -464 128 vcc FLAG -464 192 vee FLAG -32 128 vcc FLAG -32 192 vee FLAG 368 128 vcc FLAG 368 192 vee FLAG 16 464 0 FLAG 96 528 vcc FLAG 32 528 vee FLAG -2192 80 vcc FLAG -1936 -336 vcc FLAG -2192 144 vee FLAG -1328 0 vcc FLAG -2400 -288 vcc FLAG -2400 336 0 FLAG -1328 512 0 FLAG -960 256 vcc FLAG -960 320 vee FLAG -1072 336 0 FLAG -1472 592 vee FLAG -1936 512 0 FLAG -1472 0 vcc FLAG -1680 0 vcc FLAG -752 1200 0 FLAG -2512 288 0 FLAG 0 1056 vee FLAG -1680 1232 vee FLAG 144 992 0 SYMBOL res -400 16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 10K SYMBOL res -176 16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value 16K SYMBOL res 240 16 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 16K SYMBOL cap 0 16 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C1 SYMATTR Value 10n SYMBOL cap 400 16 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C2 SYMATTR Value 10n SYMBOL res -144 -224 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R4 SYMATTR Value 10K SYMBOL res -144 -80 M270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R5 SYMATTR Value 82k SYMBOL cap -800 848 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C3 SYMATTR Value 1000n SYMBOL res -288 1072 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R10 SYMATTR Value 330k SYMBOL diode -144 1072 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D1 SYMATTR Value 1N914 SYMBOL res -368 912 R0 SYMATTR InstName R11 SYMATTR Value 40.2K SYMBOL res 240 592 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R18 SYMATTR Value 10K SYMBOL res 96 432 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R19 SYMATTR Value 10k SYMBOL diode -336 656 M0 SYMATTR InstName D2 SYMATTR Value 1N914 SYMBOL diode -96 656 M0 SYMATTR InstName D12 SYMATTR Value 1N914 SYMBOL diode 48 656 R0 SYMATTR InstName D13 SYMATTR Value 1N914 SYMBOL diode 496 656 M0 SYMATTR InstName D14 SYMATTR Value 1N914 SYMBOL res -368 736 R0 SYMATTR InstName R13 SYMATTR Value 68k SYMBOL res -128 736 R0 SYMATTR InstName R14 SYMATTR Value 68k SYMBOL res 80 736 M0 SYMATTR InstName R15 SYMATTR Value 68k SYMBOL res 464 736 R0 SYMATTR InstName R16 SYMATTR Value 68k SYMBOL voltage -1936 672 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V3 SYMATTR Value 15 SYMBOL voltage -1936 816 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V4 SYMATTR Value 15 SYMBOL OpAmps\\LT1679 -464 160 R0 SYMATTR InstName U5 SYMBOL OpAmps\\LT1679 -32 160 R0 SYMATTR InstName U6 SYMBOL OpAmps\\LT1679 368 160 R0 SYMATTR InstName U8 SYMBOL res -432 272 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R8 SYMATTR Value 10k SYMBOL res -640 272 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R9 SYMATTR Value 560 SYMBOL cap -768 272 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C8 SYMATTR Value 1000n SYMBOL res -816 960 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R6 SYMATTR Value 47k SYMBOL cap -848 928 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C4 SYMATTR Value 10000n SYMBOL OpAmps\\LT1679 64 528 R90 WINDOW 3 37 22 VRight 2 SYMATTR InstName U1 SYMBOL pnp -2000 160 M180 WINDOW 3 84 0 Left 2 SYMATTR Value 2N3906 SYMATTR InstName Q5 SYMBOL res -1952 -304 R0 SYMATTR InstName R12 SYMATTR Value 56k SYMBOL res -1744 -176 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R25 SYMATTR Value 2.7Meg SYMBOL cap -2144 -16 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C7 SYMATTR Value 10p SYMBOL npn -1744 224 R0 SYMATTR InstName Q2A SYMATTR Value BCM847BS SYMBOL npn -1392 304 R0 SYMATTR InstName Q2B SYMATTR Value BCM847BS SYMBOL res -1664 576 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R28 SYMATTR Value 120k SYMBOL res -1344 32 R0 SYMATTR InstName R29 SYMATTR Value 100k SYMBOL res -2416 -48 R0 SYMATTR InstName R32 SYMATTR Value 10k SYMBOL res -2416 176 R0 SYMATTR InstName R33 SYMATTR Value 15k SYMBOL cap -1104 256 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C12 SYMATTR Value 1000n SYMBOL res -912 144 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R7 SYMATTR Value 14k SYMBOL cap -944 32 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C5 SYMATTR Value 10p SYMBOL cap -1584 -176 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C6 SYMATTR Value 100n SYMBOL res -1456 576 R180 WINDOW 0 31 76 Left 2 WINDOW 3 31 40 Left 2 SYMATTR InstName R17 SYMATTR Value 120k SYMBOL OpAmps\\LT1679 -960 288 R0 SYMATTR InstName U2 SYMBOL OpAmps\\LT1679 -816 1072 M0 SYMATTR InstName U3 SYMBOL OpAmps\\LT1679 -2192 112 R0 SYMATTR InstName U4 SYMBOL cap -2528 192 R0 SYMATTR InstName C9 SYMATTR Value 10n SYMBOL npn -1536 224 R0 SYMATTR InstName Q1A SYMATTR Value BCM847BS SYMBOL npn -1872 304 M0 SYMATTR InstName Q1B SYMATTR Value BCM847BS SYMBOL zener -1168 1088 R270 WINDOW 0 36 32 VTop 2 WINDOW 3 -4 32 VBottom 2 SYMATTR InstName D3 SYMATTR Value BZX84B13VLY SYMBOL res -1696 1088 R0 SYMATTR InstName R20 SYMATTR Value 2k SYMBOL cap 128 928 R0 SYMATTR InstName C10 SYMATTR Value 100n TEXT -2576 632 Left 2 !.tran 0 10 0 1u startup TEXT -2576 576 Left 2 !.options plotwinsize=0 numdgt=15 TEXT -2688 1104 Left 2 !.MODEL BCM847BS NPN IS = 1.822E-14 NF = 0.9932 ISE= 2.894E-16 NE = 1.4 \n+ BF = 324.4 IKF = 0.109 VAF = 82 NR =0.9931 ISC = 9.982E-12 NC = 1.763 \n+ BR = 8.29 IKR = 0.09 VAR = 17.9 RB = 10 IRB = 5E-06 RBM = 5 RE = 0.649 \n+ RC = 0.7014 CJE = 1.244E-11 VJE = 0.7579 MJE = 0.3656 TF = 4.908E-10 \n+ XTF = 9.51 VTF = 2.927 ITF = 0.3131 PTF = 0 CJC = 3.347E-12 VJC = 0.5463 \n+ MJC = 0.391 XCJC= 0.6193 TR = 9E-08 CJS = 0 VJS = 0.75 MJS = 0.333 \n+ XTB = 0 XTI = 3 EG = 1.11 FC = 0.979 TEXT -1616 696 Left 2 ;Use
I never said you did. The schematic just showed big capacitors, and in real life they would have been tantalum bead parts.
<snip>
So that's what you see as "good company". The sexually predator isn't any kind of good company.
But the better ones are thermally symmetrical, and the non-linear junctions are tamed. There's been about fifty years of painstaking development devoted to making them very nearly linear.
As Jim Thompson was given to pointing out. He also thought that his behavioral models were pretty realistic, which implies that other people could do as well
It's not hard, nor particularly expensive, to buy very good resistors. Metal film resistors can be remarkably good.
Plastic film capacitors are usually pretty good. Charge soak is mostly the slow distortion of air bubbles in the plastic film under a high - and sustained - electric field. It's not going to be a problem for a
1kHz sine wave.
It's true that I'll probably never get it built. I could lay out the circuit on my home computer, and send the layout out to turned into a printed circuit board, populated and sent back to me as loaded board. It would cost money, but I could easily afford it.
Putting a fancy output filter and a high gain output stage onto the board wouldn't make it much more expensive.
That's what spectrum analysers are for.
That's essentially what the low distortion sine wave generators are.
It doesn't make sense to you because you've never done that kind of work, and don't know much about it. Reading what Jim Williams had to say might be a start.
Edward Rawde doesn't want to know that kind of detail.
A quad is a bit of an over-kill - two pairs is all you need (and is what the schematic part numbers call up), and would make for an easier layout. A cascode NPN on top of Q2B might improve the performance, but not by much. Edwarde's first version of this circuit had eight transistors but JM got rid of half of them.
You did ask the question.
C4 at 10uF, is a pretty large capacitor,and the amplitude control loop is minimally underdamped. An E96 resistor value at R6 could probably get you closer to critical damping.
A 10uF capacitor on a chip is a big ask.
...
Show or reference the schematic where I had 8 transistors.
That's why when it's not possible to avoid it, you have two pins on your chip to connect the external capacitor. 10uF non polar shouldn't be hard.
Silly me, I should have thought of that.
This should do it:
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
16 to 20-bit DACS are available. Given a sine lookup table in ram, one could fine tune it for sub-PPM distortion. Or use a 16-bit DAC and sum in another 16-bit DAC that nulls the distortion, even the distortion of output amps or whatever.
The problem is how to measure the distortion to close the loop.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
I never said that. I'm a one-woman guy; having more than one girlfriend confuses me.
Some guys like having many lady friends; that seems to be pretty common these days. But that's not predatory if the girls cooperate.
You may as well call the women predatory.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
About 25 years ago I used a 20-bit dac followed by a 3rd order low-pass filter with a corner frequency of around 1.5kHz in a test oscillator for evaluating a 16-bit codec. The circuit was dead-bug hand wired on a sheet of FR4. I never discovered exactly how good the oscillator was, but the combined response of the test oscillator and 16-bit adc gave about 80dB SINAD. The results are in fig 4-1 on page 36 of the following pdf.
John
The problem is to find an audio spectrum analyzer with -150 dB distortion.
John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics
Having a girl friend implies some kind of social contract. At one point I was close to two different women at the same time, but neither would have been described as my girl friend - we all lived a long way apart.
Grabbing women by the pussy is predatory. If the women are too intimidated to protest it's still predatory.
Only if they initiate the physical contact. Short skirts and low necklines may signal willingness to interact, but "predation" implies reach out and grasping.
You left out dithering. Cheap twenty bit sigma-delta A/D converters have been around since 1993 - my 1996 millidegree thermostat paper relied on one. The cheap ones weren't good for 1kHz, but the hi-fi business offered faster devices.
Twenty bits is about 120dB. You don't need to block all that much of the fundamental to be able us a 20-bit ADC to directly measure the harmonics of interest.
Not necessarily.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required