A cheap very accurate LC meter with RS232 interface

inductors

this.

Interesting. It only measures at one frequency, and that frequency is the one that the components pick, so it's not necessarily good for characterizing, say, capacitive reactance at a given frequency.

But for less than $10 out of a well-stocked junk box, I don't know if I should argue.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

Cute. I still prefer my Smart Tweezers, but they cost $400!

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

Small world, Phil Rice was one of my lecturers at LTUB a decade ago :)

My final year 'major project' is still up (9 years later) on that machine, web based info-sys renders properly in modern browsers too.

inductors

this.

Another one to keep in mind for later :)

Grant.

Have you tried it with any, e.g., 0805 100nH or smaller inductors? These often have markedly difference impedances at, e.g., the VHF or UHF frequencies you'd typically use them at than the audio or LF frequencies that most testers use; I usually end up using a network analyzer to take a look at what their impedance "really" is.

Definitely a nice little device!

^^^ What I really mean here is "markedly different impedance than jwL alone would suggest" -- that is, the losses (particularly with ferrites) become significant as do parasitics as you're approaching self-resonance.

"Jan Panteltje" schreef in bericht news:i3ucfp$tsq$ snipped-for-privacy@news.albasani.net...

FAIK the original idee was from AADE (Almost All Digital Electronics). They still have a version IIB for sale and claims the copyright of his pcb design and PIC program. Look at:

for more about it.

Others build similar circuits based upon this meter and made some improvements as well. I found them in English and in German on several places on the net.

The one I build was from Phil Rice

and worked pretty well. Nevertheless measuring coils often did not gave the expected results. This is caused by the frequency the coil is measured with. Measuring using different frequencies leads to different values. Coils with a ferrite core vary more then air coils. So though the meter is very usefull to me, for accurate measuring of a coil I have to measure it with the frequency which with I want to use it.

petrus bitbyter

inductors

this.

Looks like an excellent use for all those Palm Pilots in the junk box.

On a sunny day (Wed, 11 Aug 2010 11:25:41 -0700) it happened "Joel Koltner" wrote in :

It is a good question, and I thought about it all night. I decided that a 'lossy' core should show up as a series R, so jwl + R, and was curious what would happen. So I measured some 33 uH coil, it showed 35 uH, then added 10 Ohm in series, and it showed 39 uH. LOL Now these film resistors are sometimes 'wound' with a spiral, but I dunno if this one is. Of course 10 Ohm is a ridiculous R for a coil of so few turns.

If the ferrite you are referring to causes a complex impedance, then anything can happen. So I decided to measure some small coils with ferrite adjustable cores, some from an old UHF + VHF TV tuner. It measures these no problem, and the values make sense. I do not know the exact values as these tings are adjustable anyways, but great to make some tuned circuits that are at least close to the frequency you want. I also measured some toroids, and that works OK too, but the max 10 mH is easily exceeded, I measured up to 33 mH, and that seems OK, (L should go up as square of the number of turns), but higher then 33 mH is out of range, and 1 uF is out of range too, 220 nF worked OK.

I will probably test some more coils, the thing is very useful for me, often switchers require 100 uH, 47 uH, etc, and it is very accurate in that range, you can just wind your own coils. And that is what I build it for. Also fun is that I took a 20 cm piece of wire, measured its inductance, then turned into a loop or 6, and measured it again. By pressing the loops together you can tune the inductance. You could put your lossy core in it and see what happens. Any 2 meter amateur or even if you ever build a FM or TV tuner, will recognise my 'tuning' method. :-) Finally the BFR96 transistor switch could perhaps be replaced by the proper MOSFET, this because with Vce zero and transistor off, collector floats, and the thing behaves like a varicap of a few pF, maybe a high value resistor to

• would keep collector emitter cap low, have to try some things.

On a sunny day (Thu, 12 Aug 2010 01:18:55 +0200) it happened "petrus bitbyter" wrote in :

Too bad you cannot patent software :-) Anyway I upgraded it with RS232, making it possible to for examle do a curve of a varicap on the PC aitomatically with even the soundcard output as drive voltage (or a separate ADC), so it is a new design. Also you cannot patent math. This is just basic electronics, if you want electronics in BASIC, or whatever. But theirs has a larger L range, and a larger price (129$!!!!!!!!). I just saved 110 $ LOL.

I think the RS232 option is cool, I did an RS232 frequency counter in a D connector shell too:

I built this meter up a few years ago and it is useful I did my own layout and used different components but it is cheap about ten to fifteen bucks was my total investment including a handheld enclosure to put it in. Your ? got my curiosity though. I've never used it for measuring tiny inductances lowest are beads usually they are 4.7uH. But here is a sampling.

HF INDUCTOR, 680NH, 280mA 5% 650MHZ,0805

Measured 640nH

The lowest I have is a 33nH 0603 and I cant get a reliable reading on that.

For this 220nH,1008 one I get 260nH

I tried measuring some 0603 in the 100 to 330nH but with my leads its pretty difficult. So I guess I'll have to take Digikey's and Newark word on those ;-).

I mainly use it for smps 4.7uH and up and its always given reasonable readings..

For a 27mH CMC I GET 26.1mH.

This is good enough for my needs and beats dropping a couple of grand on an LC meter.

On a sunny day (Thu, 12 Aug 2010 08:44:19 -0400) it happened Hammy wrote in :

What did you use to switch the reference capacitor, BJT? MOSFET? Relay?

I Used a reed relay.

I was going to use 2n7002 or similar but I thought the drain capacitance might effect the accuracy.

On a sunny day (Thu, 12 Aug 2010 11:44:15 -0400) it happened Hammy wrote in :

Yes, good solution, reeds, but snice I take the power from the RS232 DTR I think I am limited to less then a few mA... Maybe I could use a bipolar relay and discharge a cap into it... Have to think about this.

Thanks Jan -- that's good information!

But you're then plugging rs232 -> USB adapter? So there's 5V at ~500mA available to play with?

Pop the lid of the serial/USB bridge?

I ordered a couple serial usb bridge kits yesterday, they have the

5V and 3.3V available, as well as modem signals and shutdown option, and windoze/mac/linux drivers -- I'll report on it when it arrives.

I'm not about to waste time reinventing USB connectivity ;)

Can the reference cap be a higher value? My interest is in larger inductor values too.

Grant.

On a sunny day (Fri, 13 Aug 2010 07:59:23 +1000) it happened Grant wrote in :

Well, not from the adaptor.

But sometimes I run it simply from COM1 (/dev/ttyS0 in Linux).

These adaptors I use are really cheap from Germany, and are excellent quality.

I do not think the reference cap needs to be bigger to measure higher L values. There are 2 issues:

1) will the oscillator still run? 2) The math in in floating point asm, if you are going to change the 4MHz or the cap values, or the oscillator frequency, then you will have to change some constants, those are in Microchip 24 bit floating point format... I was actually thinking, since I have the RS232, I can make a new version that sends the raw data to the PC, and do the math on the PC in C, or even BASIC (a PIC BASIC stamp I have heard of) would be possible. You could simply have it send the 3 frequencies like this LF. RS232 is cool, maybe 2 way, so you can send to the device too, and on the PC the math is a lot easier to write. Import to mathhlab?

I think I have fixed the BJT transistor switch problem (why Rice used a relay). I added a 820k resistor from the collector of the BFR96 to the +4.7V. This is so if the transistor is off, during measurement, than the collector goes to +4.7 V in a few milliseconds (software waits a while, as it expect a relay), so the Ccb and Cce are much smaller than with the collector floating and moving up and down with the RF via the reference capacitor. The accuracy seems better now . I have put a remark on the website, so as a lot of people seem to be downloading that, please add that resistor. Then there is likely no need for relays or MOSFETS.

I've not looked at the code, It would confuse me, so close to starting coding up my latest 16F887 project (had to go to the '887 40pin chip for enough pins, at least for this prototype with own LCD module, was planning on the '886 28pin one).

the cap values,

Hmm, why is it 'they' didn't write packed BCD float point with arbitrary precision? Or doesn't the PIC have a half-carry for packed BCD? I saw the bit about keeping the value inside 16bit.

Not looked at Microchip's float stuff yet, avoided it by trimming gain in hardware for current 16bit ADC thing.

Most difficult thing I've done so far on PIC is 16bit binary to decimal to ascii.

sends

BASIC stamp I have heard of) would be possible.

the math is a lot easier to write.

I don't know mathlab, though what's in the highlight box does look a bit too much for a 'bc' one liner ;)

collector floating

Maybe Rice would have tried harder if he didn't have a relay handy?

downloading that,

Grant.

Jan Panteltje schrieb:

Hello,

if you scan your hand drawn circuit diagrams, an easily readable result would be nice. Using white papaer and a really black pen would give more contrast.

Bye

On a sunny day (Fri, 13 Aug 2010 12:11:50 +0200) it happened Uwe Hercksen wrote in :

WTF do I care. Bye