2007-04-28
"C" and "R" Jigs
How many times have you wanted a random capacitance or resistance value while prototyping? I find myself soldering in a trimmer, or clipping a capacitor across a coil to dip all the time. Once you've tuned your circuit, you end up pulling the device out and measuring it to replace with a more permanent component.
No more! I finally got around to building calibrated resistance and capacitance jigs for just these kind of "try and replace" cycles. Everyone that does RF development should have them, they are very simple but save much time. The hour or two of piecing them together will be repaid in no time.
This is hardly a new idea, no doubt most people have thought about doing this, but like myself have probably never got around to it...
"R"
For resistance I choose a 500 Ohm log pot. Log because I wanted to have a reasonable range, but still have an expanded scale at the lower-end for impedance bridge work. I've also finally found a use for those cheap "potting boxes" that Jaycar sells! (Oddly enough the chap at the counter in Jaycar York St Sydney asked me what I'd be building into them - at the time I didn't really know and said I'd think of something eventually)
That's about as simple as it comes, a pot in a box with a BNC socket. The calibrated scale is just a piece of cardboard that has been laminated and backed with adhesive using my XYLs Xyron 510 machine. This has proved to be a very good way to make dials for homebrew gear.
"C"
For capacitance a normal polyvaricon was placed in a larger potting box and some banana sockets bring out the capacitance for use (A pair of alligator clips soldered to banana jacks is also provided). A switch is offered to short the two gangs together for higher capacitances, giving three usable capacitance ranges with some overlap. The scale was then calibrated with a capacitance meter carefully nulled before use.
Making Your Own
Housing and mechanical construction I leave to the builder's tastes, but the crucial requirement is a calibrated dial with enough data points for your expected uses. The entire point of these kinds of jigs is to make them easy to read and handle mechanically, without pulling them out of the circuit and measuring with other test equipment. A reduction drive would be a good investment for the capacitance jig.
Obviously you'll need alligator clips, adapters, fly-leads, binding posts, whatever your taste/needs. Its useful to terminate the jig in something "standard" for your setup so you can share around adapters with your other test equipment.
I've found it quite handy to have binding posts and alligator clips soldered to banana jacks, and short fly-leads with alligator clips terminating in a BNC.
A BNC to clips lead with an inbuilt current balun is very handy if you are using your impedance bridge into balanced transmission lines, and is quite trivial to make: Just wind a few turns of bifilar pair around a ferrite ring, Cat5 cable is a good source of pre-twisted wire for this.
BTW: Cat5 cable is a good transmission line into the VHF region, all pairs are the same but have different twists to minimise cross talk. The Zo is around 110 Ohms, and the loss isn't too bad, especially at HF, but the wire gauge limits the power handling. I've found the velocity factor to be around 0.676 for my particular box of Belden stuff.
Uses
I originally built the "R" jig as a "known" but adjustable resistance for my Impedance Bridge. With a signal generator of some description and a pair of identical leads (Zo not critical if they are the same length), the impedance bridge becomes a lashed-up antenna analyser. It is very interesting to sweep your antennas with your generator and such a setup. Experiments with this kind of device have inspired me build a stand-alone impedance bridge (with compensated reactance measurement as well as resistance).
BTW: I use my dip meter as a signal generator in this kind of application. A BNC socket with a several turn coil soldered across it is placed over the dipper coil, this couples out enough RF to drive the Z-bridge nicely. This makes the arrangement into a kind of "super dipper" than can not only see resonances, but also tells you the load resistance at resonance. Add some variable "C" and you can measure complex impedance as well - more on that in a moment.
The "R" gadget is also handy for picking emitter resistors for a specific collector current. Sure modeling is better, and I have a calculator for this very purpose, but sometimes you just want to turn up the Ic until something bad happens (instability, smoke, etc), then back it off - the "R" jig makes this an easy exercise. This particular use has also inspired some thought about building an "RF gain" testing jig for transistors. A scale calibrated in Ic would let one quickly check the optimum Ic for gain for the particular device, especially when looking to optimise noise performance. If nothing else it would confirm your random junkbox "TUN" or "TUP" is actually an "RF transistor" and let you pick "hot" ones when you need them.
Finding the optimum termination resistance for a crystal filter for best flatness is another use of such a jig. A larger pot, 1-2k might be more useful in such an application but you can always pad out a smaller value with fixed resistances.
A dual-gang pot that tracked sufficiently well could be used to implement a "Zo finder". A swept (or switched non-harmonically related) frequency generator could feed a bridge using one of the pots as the "known", and the other gang could terminate a length of tranmission line under test. The pot's resistances would match Zo when best "flatness" was achieved across the frequencies injected. This would make the device something you can "twist the knob until the meter scale wiggles the least - then read the Zo"... (This is how I "measured" the Zo of cat5 BTW, but it is also possible to determine it by measuring the RLC properties of the line directly - however for accuracy this means a long length or very good gear, the flatness estimate can use a few metres)
For the capacitance jig the most obvious use is finding the "C" needed to resonate with something before replacement with a fixed value. It is also handy for dipping random inductors, but this means you need to do some math (or use my resonance calculator), unlike a fixed value for which you can construct a table or simple nomogram to make the operation more expedient.
I have tried using the "C" jig as a reactive measuring element with my Impedance Bridge. A "mid-scale" capacitance is placed in the unknown arm of the bridge to "center" the scale as R+j0, and capacitances relative to this middle value can be converted into +ve or -ve reactances at the particular frequency of interest. My reactance calculator can be helpful there. For best accuracy a carefully constructed device with compensation inductance is really needed, but the lash-up will work nicely on lower bands, (calibration on spot-frequencies works too).
For picking bandset and bandspread capacitances it is very expedient. However I've recently written a special VFO Helper calculator to do this analytically for you. It is a pretty handy too actually, letting you calculate two capacitance values that will give you the desired set and spread for any VFO. I used it design my dipper coils for the best coverage given my particular variable capacitance range measured at the coil socket.
"L"?
Will there ever be an "L" to add to the family? Very likely yes! There are several ways to implement this, but the hard part is fitting a dial to them. A roller inductor with counter would be ideal, but expensive and requires a table to match turns to inductance. A slug is hard to accurately track unless I couple it to a turns counter, and that means some kind of table too (although the old brass screw idea would work and be pretty easy to mark in depth of penetration). A pancake coil would be fairly easy to mark calibrations on and build mechanically, as would a conventional solenoid and an alligator clip lead to tap it, but both have uncontained fields and connection noise problems to deal with. A similar arrangement with a toroid would be hard to mark and use unless it was very large physically. A variometer is probably the easiest to homebrew and needs no sliding contact. I'll probably write a variometer calculator to design them first, so I can understand the geometry trade-offs. Until then I have a collection of calibrated inductors I made for the L meter project which I can use in a less expedient way (or use them to build a 'decade' box).
2007-06-10 Update
I put together a lower value "C"-jig for my VHF work on the Fredbox. I am building its TX stage again using fixed capacitances, and this little tool was helpful in picking the capacitors that would resonate mid-range of the toroids "stretch and squeeze" inductance range.
It is a "red" trimmer on a small fragment of phenolic matrix board, with two 0.71 mm tinned copper wires stitched into the holes as stress relief. The label was made in the same manner as the other jigs, with the XYLs Xyron machine using its laminate/adhesive cartridge.
I also raided Tanya's jewelery making supplies for the pointer. It is a piece of silver plated nickel memory alloy wire normally used in "wine glass rings", those things that people use at parties to distinguish their beverage glasses with unique beads or other charm things on them.
2 comments.