Clifton Laboratories 7236 Clifton Road  Clifton VA 20124 tel: (703) 830 0368 fax: (703) 830 0711

E-mail: Jack.Smith@cliftonlaboratories.com
 

I've previously designed and made available a 200 KHz wide 4912 KHz bandpass filter as part of the Z90/91. I've had questions about a similar filter for the K3, but wider, to be used as an additional protective measure between the K3's IF output and a Softrock receiver.

I've today build a prototype filter for the K3's IF, 8215 KHz. The specifications are:

• Bandwidth (-3 dB points): 400 KHz nominal
• Design form: Butterworth
• Passband loss: 4 db nominal
• Input/Output impedance: 50 ohms

Physically, the 8215 KHz filter is identical with the K2 bandpass filter, and photographs can be seen by clicking here, or via the K2 Filter link at the left.

I'm reluctant to make these filters available as a kit, as they require the four toroid inductors to be wound and adjusted to the target inductance within ±2%. Even with the inductors within this tolerance, final adjustment requires adjusting four trimmer capacitors to provide the desired center frequency and passband shape. I use an HP4342A Q-meter to adjust the inductors and verify their Q as meeting minimum standards, and an HP8752B vector network analyzer to tune the filter. 

The original K2 filters were supplied as an assembled, tuned and tested assembly, with a tamper seal on the enclosure. The problem with this approach is that it takes an hour and a half for me to wind the inductors, adjust them to target values, coat them with Q-dope, assemble the PCB, drill the enclosure, tune the filter in a special enclosure and then move the tuned filter assembly to the normal enclosure.  Considering the cost of material and time, this means the price on a stand-alone, assembled and aligned filter is going to have to be in the $45 range.

The results are decent, meeting my specifications, as seen in the passband sweeps shown below. The asymmetry in the passband response is inherent in coupled resonator filters, as the frequency increases, the filter devolves into a series/parallel capacitor network.

If there is sufficient demand, I will consider a run of both the K2 bandpass filter and  the K3 variant. I will also consider a custom designed filter at additional cost for IFs in the 4-12 MHz range.