Reducing attenuator loss using ferrite beads

Created: Nov. 2022

 ------------------------

Introduction

The 4-diode PIN attenuator was originally designed to operate with a 0 - 15V control voltage [1 - 2]. Later, with the advent of low voltage electronics, the 15V requirement became infeasible. When the attenuator’s bias network is modified to enable a reduced 0 - 5V control range, the trade-off is higher insertion loss (minimum attenuation); i.e  at 10 MHz, increasing from the original 3 dB to 5 dB [3 - 4].

 Interest in reducing the insertion loss came mostly from makers of cable TV distribution amps. As the attenuator followed the amplifier, a lower loss in the former would permit the use a smaller transistor in the latter's final stage; i.e. a cost saving. Similarly, if the attenuator is part of a receiver front-end, reducing the insertion loss can beneficially improve the overall noise figure.

 The attenuator’s bias resistors contribute to the loss by dissipating the RF energy as heat. So, one way to reduce the insertion loss is by adding inductors in series to the resistors, but the resultant improvement is narrowband [5]. Alternatively, replacing the inductors with ferrite beads can enable wideband improvement over several octaves [6]. The bead-equipped attenuator was previously published without circuit simulation. So, this article plugs the gap by demonstrating the simulation and its results.

 Material & method

The attenuator has 3 resistors to control the bias current through the 4 diodes. As the resistors are connected in shunt with the RF path, they dissipate part of the RF signal. The loss due to the bias resistors can be ascertained using simulation (fig. 1). Initially, the 330R bias resistor R1 is simulated alone without the ferrite bead  L2.  The simulated loss is ~0.6 dB (blue trace). Subsequently a model for the Murata BLM18RK102 bead is added to the simulation circuit. Adding the ferrite bead in series with the resistor, reduces the loss to ~0.2 dB (red trace). The combination of bead and resistor results in a peak at ~180 MHz. This peak is due to the bead's limited inductance increasing loss at the lowest freq. A higher inductance bead could have flattened the low end response, but I just used whatever bead was available.

Fig. 1  Adding a ferrite bead (L2) in series with the resistor (R1), reduces the insertion loss from 0.5 dB (blue trace) to ~0.2 dB (red trace). 

In the experimental prototype, BLM18RK102 beads were added in series with the 3 bias resistors R1-R3 (fig. 2).

The attenuator requires a constant voltage, V+, and a variable control voltage, Vc (fig. 2). With V+ equal to 1.5 V, the variable control voltage will range from 0 V to about 5 V.

The HSMP-3816 is a diode quad housed in a five-pin, leadfree SOT-25 surface mount package. The pi-connected quad PIN diodes are adjacent die selected from the same wafer for closely matched electrical characteristics. In addition to the obvious size advantage, bundling four well-matched PIN diodes into one SOT-25 package helps ensure better symmetry between the attenuator's input and output arms than is realizable using physically distinct parts.

Fig. 2 PIN diode attenuator schematic diagram with added ferrite beads


Table 1 Part-list

fig. 3 Attenuator PCB layout showing modifications to fit ferrite beads

Fig. 4 component placement diagram

The attenuator was implemented on an FR-4 printed-circuit board (PCB) of 0.8 mm thickness (fig. 5). However, this low cost material curtails the attenuator's high frequency performances by increasing loss.

Fig. 5. Photo of assembled attenuator showing its compact form & low component count

PIN diodes cannot be modeled on SPICE because the standard PN diode model has no provision for minority carrier lifetime; i.e. this parameter permits the PIN diode to behave like a fixed resistor at RF. A workaround is to model the PIN diode using a linear equation [7] as per implemented. In the APLAC simulator. The HSMP-381x's model parameters were obtained here [8].

The equivalent circuit model has a three-level hierarchy: APLAC linear model of diode, packaged diode array and complete attenuator (fig. 6). For convenience, the RF connectors are not modeled. The ferrite bead model is described in more detail in another page on this site. 

Fig. 6 Equivalent circuit model. See appendix for detailed views

Results

 

Both simulated and measured insertion losses are ~2 dB at mid-band (fig. 7). The simulated result’s error is < 0.5 dB. The improvement in insertion loss is ~3 dB compared to without beads [3 - 4]. The  insertion loss is lowest at 100 - 250 MHz. It may be possible to shift this frequency range by carefully selecting different ferrite beads, although this has not been tried.

Fig. 7 Simulated & measured insertion loss vs frequency (Vc = 5V). Simulation achieved almost perfect agreement with measurement. The loss reduction is almost 3 dB at VHF

At 100 MHz, the attenuation can be varied from 55 dB to 2 dB (fig. 8). Both simulation and measurement agree almost perfectly.

While the insertion loss (minimum attenuation) is lowered by these modifications, the maximum attenuation (corresponding to the lower limit of the control voltage range) remains the same. Therefore, the reduction of the attenuator’s loss also provides the additional benefit of improved dynamic range (the difference between minimum and maximum attenuation). On the flip side is the increase in the rate of change for the attenuation versus control voltage curve (fig. 8 & 9). Most the attenuation range is now concentrated around the 1 to 2 V region.

Fig. 8 Simulated attenuation versus control voltage (Vc) at 100 MHz. Simulation (red dash) agrees with measurement (solid black)

At 500 MHz, the attenuation can be varied from 41 dB to 2.5 dB (fig. 9). Both simulation and measurement agree almost perfectly.

Fig. 9 Simulated attenuation versus control voltage (Vc) at 500 MHz. Simulation (red dash) agrees with measurement (solid black).

Appendix

Fig. 10 Equivalent circuit model of PIN diode chip

Fig. 11 Equivalent circuit model of HSMP-3816 diode array

Fig. 12 Equivalent circuit model of attenuator

“I am never content until I have constructed a mechanical model of the subject I am studying. If I succeed in making one, I understand, otherwise I do not.” - Lord Kelvin

References

[1] Waugh, R. W., " A Low Cost Surface Surface Mount PIN Diode PI Attenuator," Microwave J., vol. 35, no. 5, May 1992, pp 280-284.

[2] Hewlett Packard Application Note 1048, “A Low Cost Surface Mount PIN Diode π Attenuator”, 1996. Available: http://www.hp.woodshot.com/hprfhelp/4_downld/lit/diodelit/an1048.pdf

[3] C. L. Lim, S.C. Goh & Y.C. Lim, "Diode quad is foundation for PIN diode attenuator," Microwaves & RF, May 2006. Available: http://mwrf.com/semiconductor/diode-quad-foundation-pin-diode-attenuator

[4] “PIN diode attenuator simulation”, Nov. 2022. Available: https://sites.google.com/site/randomwok/Home/electronic-projects/pin-diode-attenuator-simulation

[5] L. F. Fei, "A low Cost, Compact, Pi-Configured PIN Diode VCA," Applied Microwave & wireless, Nov. 2000.

[6] C. L. Lim, "Cut Loss in Low-Voltage, Wideband PIN Attenuators", Microwaves & RF, Apr. 2008.

[7] Joe Walston, “SPICE Circuit Yields Recipe for PIN Diode”, Microwaves and RF, pp. 78-89, Nov., 1992.

[8] "SPICE Library: PIN Diode Models", Available: http://www.hp.woodshot.com/hprfhelp/design/SPICE/pins.htm