microwave freq. response of axial lead thin film resistor?

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Does anyone know where I can find some basic info on this? I have to use this horrible part, 100-ohm 5%, as a 'lumped element' in a microstrip circuit (power divider) for a school lab experiment, though it'll be anything but lumped. Having 15mm+ leads, soldering to microstrip stubs, and it being a thin film resistor (inductive trim lines) got me to make up this model:

     solder                 C_series                    solder
     joint              |-------||--------|             joint
------???---nnnn--------|                 |---nnnn-------???--------
    /      L_stub1  |   |--nnnn----/\/\/--|  L_stub2  |       \
 Z0 \      C_stub1 ===    L_Rtrim    R       C_stub2 ===      / Z0
    /               |                                 |       \
--------------------------------------------------------------------

The solder joint effect is unknown (some C and L to GND) and probably not reproducible, so it's the first thing I'd like to ignore. Next, the axial leads were treated as transmission lines, but they are each only 10% of a wavelength, and symmetric, which isn't a great argument but one I'll use to ignore them, anyway. The series capacitance is likely tiny and leads to a very high frequency pole, so it disappears. The Z0's are the microstrips (characteristic impedance).

Leaving this:

---------nnnn----/\/\/---------
    /    L_Rtrim    R     \
 Z0 \                     / Z0
    /                     \
-------------------------------

Does it look familiar, or have I simplified it too much? Do you know what to expect for (trimmed helical thin film in resistor) inductance?

Even my microwave book (Pozar) doesn't have this crud... must be too close to 'real life'.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

However you model the equivalent circuit it will be correct at only narrow band range of frequencies ( due to distributed nature of capacitance/inductance/ resistance.

Microwave on the other hand is a broad statement. Can You be a bit more specific about the range of frequencies of interest?

Meanwhile 100 ohm metal film resistors are OK for use up to approx 1 GHz. Another broad generalisation. If you can find carbon composition resistors so much the better.

You can trim metal film resistors for broadband operation by including some adjustable shunt capacitance ( in the form of a deformable foil capacitance either to ground or in paralele with resistor body.

The resistor leads contribute approx 10 nH per centimetre of lead length. Try to design the microstripline components to minimise the efects of lead inductance.
Alternatively use hybrid design ( transformer based ) for a broadband performance.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There is a "sweet spot", resistance wise, for high frequency operation of axial film resistors. My recollection is that it is sort of in the 50-200 ohm range.

Higher resistances are created by cutting a spiral in the film on the cylindrical body of the resistor. Somewhere in the low-hundred ohm area, depending on the film resistivity, the manufacturer does NOT cut the spiral.

At lower resistances, I remember being told, though I don't recall the mechanism, that endcap-endcap capacitance becomes the dominant reactive element.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We're supposed to divide a 1 GHz signal. The leads are about 17 mm each, giving 5.7 nH each in air, though I suppose the ground plane, about 1.9 mm below, will increase this. (The Wikipedia formula gives me ~800uH, which is bogus.) Thanks for the idea, ig - I could just stick on some solder wire 'rabbit ears' to the leads and adjust their spacing for shunt capacitance control.

That makes sense, Jim. Maybe the net effect will be 2nd-order lowpass, then, with the spike above 1 GHz.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The ground plane may provide the necesary resistor body/leads capacitance to ground in order to provide the mnecesary impedance for the spliter.

I used to use thin magnet wire and metal chasis to achieve transmission line effect in wiring the hybrid transformers ( couplers and splitters in MATV/CATV components).

Used to load window putty with metal filings and use the mixture to dampen resonances in some cases.

Do You have access to a return loss bridge? or a network analyser? You can build the bridge quite readily and with some care obtain an instrument capable of measuring return loss down to 30dB at VHF and better than 20 dB at UHF( say 800+MHz)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yep, I'll be tuning this with a 5 GHz VNA, but I only get about 2 hours.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Tuning?There is not much to tune in the sense that You can not alter line widths and lengths.How will You launch the signal? SMA connectors? How will you mount them?
Will You use a single resistor or may be two in parallel?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We can use as many 100 ohm resistors as we want, but only those. We can alter electrical lengths, trace impedances, and reflective corners a bit by smudging solder around. Yes, SMA connectors to the VNA.