tuning an LC

done so for years. The parts are still stocked by some broad line distributors, including Newark and Farnell. I've posted part numbers in this thread.

Right, why make it more complicated than it needs to be. Can't get much simpler than an adjustable pot core and it's cheap too.

The varactor solution is elegant and can be tuned in software, but then you need to design and debug the software to drive it and tune the control loop. All costs time & money, even if it might be an interesting exercise in itself :-)...

Chris

Does anybody sell assembled adjustable pot cores around 77n uH? If not, I'd have to...

Find the parts: core halves, bobbin, adjuster, header with pins, any additional mounting hardware, wire.

Create a database entry and stockroom bin for each part

Buy them

Figure out the winding and such and make an assembly drawing and BOM. Check and release such drawing.

I've signed NDAs, all the emails about the system are encrypted, and I'm not supposed to know what the end application is. We provide an LC that simulates some LC, digitize the voltage across it, and do a bunch of DSP in an FPGA and one ARM core, and output something. It's an aerospace application. Sounds like an LVDT, vaguely, or some sort of inductive position sensor.

Not melting plastic, I suspect.

It's a secret... you're supposed to suggest a solution to the secret problem >:-} ...Jim Thompson

You didn't properly guess my _secret_ solution ;-)

...Jim Thompson

Wasn't interested enough to look at the simulation. When the math is that easy, why use SPICE? Particularly when the point is clearly a putdown rather than being helpful?

Cheers

Phil Hobbs

Based on the "information" provided... produce 60V P-P at 107kHz, I solved it with no critical tuning required.

Now it's slowly coming out, in typical Larkinesque fashion, that is not quite what is needed.

Who really knows... ...Jim Thompson

English is my first language, but that doesn't often show. What I meant is the Capacitance vs time, voltage function. I have a lot of respect for Taiyo-Yuden, met the owner's son once in Sunnyvale [if 'like father like son'; the owner was probably a good man, too] TY puts a LOT in small packages, but have to admit that drop in C vs voltage is a bit harsh, ...and unexpected.

However, you have an interesting point. 107kHz. Aerospace. Secret.

Hmmmm that could be submarine communications. Probably not for 'down hole' monitoring to punch through salt layers.

How about 'wireless power transfer' ?? You can get a lot of power transferred using 107kHz, not hurt people, and not heat up their wedding rings, etc. The fact that I was specifically forbidden to use that frequency band and that Aerospace is involved tilts me towards submarine communications, or more likely underwater robotic vehicle communications, UUV communication.

If it's for 'position sensing' or some type of 'magical locator'; now that's just stupid.

On a sunny day (Tue, 19 Nov 2013 15:46:42 GMT) it happened Jan Panteltje wrote in :

google ultrasonic airplane wings About 816,000 results (0.18 seconds) about detecting ice on wings and structural integrity... etc etc

Maybe stealth too (vibrate and diffuse radar?) Think I did see that once...

I am back into this:

May or may not interest you, long lecture Hidden Variables in Modern Cosmology (Antony Valentini) (professor). I really enjoyed that. Always been a De Broglie fan... :-)

I posted the Spice model of the LC and its driver on 11/14. I've stated several times that my specific problem is to tune the LC, in production, to +-2% of the drive frequency.

Having trouble understanding that?

We're all dying to know about yet another of your brilliant but secret circuits. The sine wave that you posted does indeed look like a sine wave. Fascinating.

done so for years. The parts are still stocked by some broad line distributors, including Newark and Farnell. I've posted part numbers in this thread.

simpler than an adjustable pot core and it's cheap too.

Earlier in this thread John Fields suggested a 50uH +/-10% fixed inductor a nd a trimmable inductor from the same manufacturer to pad it up to 77uH.

tional mounting hardware, wire.

I've told you were you could find the adjuster, and given you the URL for t he EPCOS P26x16 components, which lists everything (with part numbers) exce pt the enamelled wire (which you can buy from Newark, but probably not at t he exactly optimal gauge)

One half core, one former, one mounting plate, one adjuster and one sort of wire. A total of five. Your 8-4-2-1 capacitor trim scheme would have neede d four bins for the capacitors alone, not that you can buy capacitor values that fit that sequence.

Half an hour. Allow a bit longer if you want to find Litz wire.

ck and release such drawing.

The winding isn't exactly difficult to figure out. I've already given you t he number of turns (22) for a 160nH per turn squared pot core (which is wha t you'd most likely buy). I'd be surprised if even you haven't already got an assembly drawing for a pot core somewhere in your system, but you could always adapt the relevant drawing from the EPCOS data sheet.

And what you originally proposed needed a multi-stage trimming operation to get it into tolerance. Telling the production how to twiddle an adjuster i s less demanding.

You do have to check that you drive levels won't get your core close to sat uration. My guess is that it wouldn't, but I haven't bothered to explicitly check - you do need to learn a bit more about inductors, so I'll leave tha t to you, but do note that LTSpice does support the John Chan model for hys teretic inductors.

Well, not unexpected, it's normal for ceramic caps. Y5R is very crappy, X5R better.

The function is C versus bias voltage

If you want manufacturer data, Murata has a lot of good details at:

This is an excellent graph:

Cheers

Klaus

That's way too optimistic. Murata's datasheets show 55-70% capacitance decrease for X7Rs at full voltage. This really bites hard for regulator output capacitors.

Example of a 10uF, 10V, X7R cap:

On Tue, 19 Nov 2013 13:56:59 -0800 (PST), Bill Sloman wrote:

--- . +30 . | . +-------+ . | | . [L1] | . | [C1] . [L2] | . | | . +-------+ . | . [100R] . | . D .PWM>----G . S . | . GND

L1 = 56µH +/- 10% Caddell-Burns 7500-10 L2 = 18-33µH Caddell-Burns 6740-30 C1 = 27nF +/- 2%

. L1 high L2 . +--61.6µH + 18.7µH-->f1 . C1 high | . 27.54nF----+ . | L1 low L2 . +--50.4µH + 29.9µH-->f2 . . . L1 high L2 . +--61.6µH + 21.9µH-->f3 . C1 low | . 26.46nF----+ . | L1 low L2 . +--50.4µH + 33.2µH-->f4 . . . 1 . fn = ---------------------- . 2pi sqrt((L1+L2) C1)

Version 4 SHEET 1 1312 756 WIRE -432 64 -512 64 WIRE -256 64 -432 64 WIRE 48 64 -32 64 WIRE 224 64 48 64 WIRE 544 64 464 64 WIRE 720 64 544 64 WIRE 1024 64 944 64 WIRE 1200 64 1024 64 WIRE -432 112 -432 64 WIRE 48 112 48 64 WIRE 544 112 544 64 WIRE 1024 112 1024 64 WIRE -432 224 -432 192 WIRE 48 224 48 192 WIRE 544 224 544 192 WIRE 1024 224 1024 192 WIRE -512 240 -512 64 WIRE -32 240 -32 64 WIRE 464 240 464 64 WIRE 944 240 944 64 WIRE -256 272 -256 64 WIRE 224 272 224 64 WIRE 720 272 720 64 WIRE 1200 272 1200 64 WIRE -432 336 -432 304 WIRE 48 336 48 304 WIRE 544 336 544 304 WIRE 1024 336 1024 304 WIRE -256 416 -256 352 WIRE 224 416 224 352 WIRE 720 416 720 352 WIRE 1200 416 1200 352 WIRE -512 448 -512 304 WIRE -432 448 -432 416 WIRE -432 448 -512 448 WIRE -32 448 -32 304 WIRE 48 448 48 416 WIRE 48 448 -32 448 WIRE 464 448 464 304 WIRE 544 448 544 416 WIRE 544 448 464 448 WIRE 944 448 944 304 WIRE 1024 448 1024 416 WIRE 1024 448 944 448 WIRE -304 496 -384 496 WIRE 176 496 96 496 WIRE 672 496 592 496 WIRE 1152 496 1072 496 WIRE -512 544 -512 448 WIRE -384 544 -384 496 WIRE -32 544 -32 448 WIRE 96 544 96 496 WIRE 464 544 464 448 WIRE 592 544 592 496 WIRE 944 544 944 448 WIRE 1072 544 1072 496 WIRE -512 656 -512 624 WIRE -384 656 -384 624 WIRE -384 656 -512 656 WIRE -256 656 -256 512 WIRE -256 656 -384 656 WIRE -32 656 -32 624 WIRE 96 656 96 624 WIRE 96 656 -32 656 WIRE 224 656 224 512 WIRE 224 656 96 656 WIRE 464 656 464 624 WIRE 592 656 592 624 WIRE 592 656 464 656 WIRE 720 656 720 512 WIRE 720 656 592 656 WIRE 944 656 944 624 WIRE 1072 656 1072 624 WIRE 1072 656 944 656 WIRE 1200 656 1200 512 WIRE 1200 656 1072 656 WIRE -512 736 -512 656 WIRE -32 736 -32 656 WIRE 464 736 464 656 WIRE 944 736 944 656 FLAG -512 736 0 FLAG -32 736 0 FLAG 464 736 0 FLAG 944 736 0 SYMBOL ind -448 432 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L1 SYMATTR Value 61.6µ SYMBOL cap -528 304 M180 WINDOW 0 -24 62 Left 2 WINDOW 3 -63 6 Left 2 SYMATTR InstName C1 SYMATTR Value 27.54n SYMBOL res -272 256 R0 SYMATTR InstName R1 SYMATTR Value 100 SYMBOL voltage -512 528 R0 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 30 SYMATTR Value2 AC 1 SYMBOL nmos -304 416 R0 SYMATTR InstName M1 SYMATTR Value BSS123 SYMBOL voltage -384 528 R0 WINDOW 3 24 96 Invisible 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(0 15 0 100n 100n 2.336u 9.346u) SYMATTR InstName V2 SYMBOL ind -448 208 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L2 SYMATTR Value 18.7µ SYMBOL res -448 208 R0 SYMATTR InstName R2 SYMATTR Value 2.364 SYMBOL ind 32 432 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L3 SYMATTR Value 50.4µ SYMBOL cap -48 304 M180 WINDOW 0 -24 62 Left 2 WINDOW 3 -63 6 Left 2 SYMATTR InstName C2 SYMATTR Value 27.54n SYMBOL res 208 256 R0 SYMATTR InstName R3 SYMATTR Value 100 SYMBOL voltage -32 528 R0 WINDOW 39 0 0 Left 2 SYMATTR InstName V3 SYMATTR Value 30 SYMATTR Value2 AC 1 SYMBOL nmos 176 416 R0 SYMATTR InstName M2 SYMATTR Value BSS123 SYMBOL voltage 96 528 R0 WINDOW 3 24 96 Invisible 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(0 15 0 100n 100n 2.336u 9.346u) SYMATTR InstName V4 SYMBOL ind 32 208 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L4 SYMATTR Value 29.9µ SYMBOL res 32 208 R0 SYMATTR InstName R4 SYMATTR Value 2.364 SYMBOL ind 528 432 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L5 SYMATTR Value 61.6µ SYMBOL cap 448 304 M180 WINDOW 0 -24 62 Left 2 WINDOW 3 -63 6 Left 2 SYMATTR InstName C3 SYMATTR Value 26.46n SYMBOL res 704 256 R0 SYMATTR InstName R5 SYMATTR Value 100 SYMBOL voltage 464 528 R0 WINDOW 39 0 0 Left 2 SYMATTR InstName V5 SYMATTR Value 30 SYMATTR Value2 AC 1 SYMBOL nmos 672 416 R0 SYMATTR InstName M3 SYMATTR Value BSS123 SYMBOL voltage 592 528 R0 WINDOW 3 24 96 Invisible 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(0 15 0 100n 100n 2.336u 9.346u) SYMATTR InstName V6 SYMBOL ind 528 208 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L6 SYMATTR Value 21.9µ SYMBOL res 528 208 R0 SYMATTR InstName R6 SYMATTR Value 2.364 SYMBOL ind 1008 432 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L7 SYMATTR Value 50.4µ SYMBOL cap 928 304 M180 WINDOW 0 -24 62 Left 2 WINDOW 3 -63 6 Left 2 SYMATTR InstName C4 SYMATTR Value 26.46n SYMBOL res 1184 256 R0 SYMATTR InstName R7 SYMATTR Value 100 SYMBOL voltage 944 528 R0 WINDOW 39 0 0 Left 2 SYMATTR InstName V7 SYMATTR Value 30 SYMATTR Value2 AC 1 SYMBOL nmos 1152 416 R0 SYMATTR InstName M4 SYMATTR Value BSS123 SYMBOL voltage 1072 528 R0 WINDOW 3 24 96 Invisible 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(0 15 0 100n 100n 2.336u 9.346u) SYMATTR InstName V8 SYMBOL ind 1008 208 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L8 SYMATTR Value 33.2µ SYMBOL res 1008 208 R0 SYMATTR InstName R8 SYMATTR Value 2.364 TEXT -496 680 Left 2 !.tran 1ms TEXT -592 328 Left 2 ;HIGH TEXT -408 424 Left 2 ;HIGH TEXT -112 328 Left 2 ;HIGH TEXT 72 424 Left 2 ;LOW TEXT 392 320 Left 2 ;LOW TEXT 568 424 Left 2 ;HIGH TEXT 872 328 Left 2 ;LOW TEXT 1048 424 Left 2 ;LOW TEXT -408 200 Left 2 ;MIN TEXT 1048 192 Left 2 ;MAX TEXT -496 712 Left 2 ;John Fields, 19 Nov 2013

Q's a little low and circulating current is a little high, but other than that, Mrs. Kennedy, how did you enjoy the parade?

Unfortunately, L2 has too much DCR and can't handle the circulating current.

Thanks, especially for the EXACT URLs.

The trimmable parts really are too small to cope with the currents involved, even if you start off with a +/-3% part you've got by winding a coil onto a heavily gapped pot core. You can come closer, but not close enough.

Buy the right EPCOS P26x16, wind the coil onto the former and use the EPCOS adjuster.

offers you two alternatives ex-stock for the core itself. 495-5809-ND is cheaper and you probably don't need the high resistivity ferrite on offer in the more expensive alternative.

The DCR is too high, too... blows the customer spec. I tried a Coilcraft part, same problem.