January 10, 2012

Join the "CWTD Yahoo Group" for email discussion in between our weekly sessions by clicking here.


Homebrew Filter Design and Measurement Techniques

Now in the third-and-final week of our LPF Design Mini-Tutorial, we finish things up by actually taking some careful measurements of the LPF we designed and built in previous sessions.

For those who would like to actively participate with us, our Notes (below) illustrate how we made the measurements and plotted the results.

This week we discuss these "empirical" data results, as compared with the theoretical plots in previous weeks.  And we wrap things up with a quick discussion of "Key Concepts".

~George N2APB & Joe N2CX

Audio Recording ... (Listen to MP3 recording)

Attendees ...
George N2APB, Joe N2CX, Bill W7WMB, Bob WB6KWT, Bob K7HBG, Bruce N1RX, Cecil K5NWA, Charles WC5MC, Chris W0ANM, Dave AD7JT, Dave G1OYG, David KB4FXC, Eldon WA0UWH, Frank N3PUU, G3OTH, Larry W2HHV, Joe WA2QZP, John K5JHF, John KC6CVB, Ken VA3KMD, Paul AK1P, Paul WA0RSE, Peter SV0XAW, Randy WB0SMX, Steve K1RF, Terry WB4JFI

Text Log from this week's session ...
<20:04:17>"George - N2APB": Hi everyone, this is the text are of the QRP Homebrewing channel.  We can type messages, liks and commentary to everyone on the channel using the text area at the bottom of the screen.  Please be sure to follow along with the discussion using the notes on ouor web page at http://www.njqrp.org/teamspeak/Jan%2010.html .
<20:12:07>"George - N2APB": Set up #1 that Joe's talking about is shown on our web page at http://www.njqrp.org/teamspeak/Jan%2010.html .
<20:13:52>"George - N2APB": Here's a phopt of the precision RF detector that Joe's talking about ... http://www.njqrp.org/teamspeak/FilterDesign/pics/SIC(800).JPG   ... we call it the Sweeper Input Card, or "SIC" ... soon to be an easy-but-useful kit for homebrewers & experimenters.
<20:14:24>"George - N2APB": 1 dB = 0.024 mV change in the detector output.
<20:15:58>"Bruce - N1RX": What's the resolution (in mV) of the SIC?
<20:16:27>"George - N2APB": The SIC detector provides an analog output of that DC level, between 0 and 2.7V.  But there's also an A-to-D converter (12-bits/sample) that a microcontroller can easily read over an i2c bus for subsequent computations.
<20:18:38>"Terry WB4JFI": more like a poor-mans network analyzer
<20:18:52>"George - N2APB": Yeah, you're right!
<20:19:24>"Bruce - N1RX": True. S21 measurements on a 2-port device
<20:19:51>"Terry WB4JFI": or a spec an & tracking generator
<20:21:03>"Bruce - N1RX": http://www.ka7exm.net/pic_power/index.htm
<20:21:33>"Joe - N2CX": KA7EXM PIC Power Meter site: http://www.ka7exm.net/pic_power/index.htm
<20:24:58>"Joe - N2CX": Almost All Digital Electronics AADE http://www.aade.com/lcmeter.htm
<20:25:45>"Bruce - N1RX": Core permeability varies, and turns spacing affects total inductance significantly.
<20:27:28>"Joe - N2CX": While it says RF Volts it really should be DCout from the SIC
<20:28:58>"Joe - N2CX": Also with toroidal inductors there is capacitance between turns of the winding.  Moving them on the core closer or farther apart changes the capacitance, thus the apparent inductance.
<20:29:40>"Bruce - N1RX": Joe- did these readings take into account the natural roll-off of the DDS output? That would add to the apparent "loss" through the filter.
<20:31:06>"Joe - N2CX": Bruce, no we did not take that into account for that.  I am going to address it soon in terms of setup calibration.
<20:31:31>"Paul - wa0rse": Excellent, Joe!
<20:33:44>"Bruce - N1RX": FB. As George said at the beginning, know the limitations of yur measurements 8-)
<20:45:55>"Paul - wa0rse": I wasn't able to make the first 2 T-S session. What does the future of the SIC for the 908 look like?
<20:48:31>"George - N2APB": The SIC will be an add-on card for the Micro908, coming along when we have the next round of Micro908 Kits available at month-end.
<20:49:07>"Paul - wa0rse": Thanks, George. I'll look for it.
<20:56:16>"Paul - wa0rse": Will my first-run AA-908/Micro-908 be able to accommodate the SIC with updates of its software?
<20:56:30>"Bruce - N1RX": yes.
<21:02:30>"Bruce - N1RX": Google Thevenin Theorem to read about maximum power transfer principle
<21:14:29>"George - N2APB": If interested, "refresh" your browsers that are looking at our web page ... I corrected the column headings in the measured data charts.
<21:15:58>"Paul - wa0rse": VNA
<21:18:43>"Paul - wa0rse": Thank you, George and Joe!
<21:18:59>"Frank N3PUU": thanks guys, great presentation!
<21:19:09>"Steve, K1RF": Great tutorial!!
<21:19:24>"Terry WB4JFI": Thanks again George and Joe
<21:25:55>"Peter - SV0XAW": Thanks George and Joe enjoyed all 3 sesions! learned alot. 73


Tutorial Notes ...

Measuring the LPF performance:


The approach we used for measuring the LPF performance is to inject frequencies into the input port of the filter and measure the output port. 





Set-up #1 Set-up #2
A Micro908 Antenna Analyst in "VFO" mode is used to deliver a stable, adjustable DDS-based signal to the input of the LPF.  For measuring the filter output, the precision RF detector (the "Sweeper Input Card") is integrated into the Micro908 to read the log amp output and display the result on the LCD.  An HP8640B signal generator delivers a stable, 0 dBm RF signal to the input of the LPF.  For measuring the filter output this time, we're using the precision KA7EXM "Microwattmeter".

(Click on photo for larger view)

(Click on photo for larger view)
The Results
Plot #1:  Full Sweep Plot #2:  Close-in at Roll-Off
Theoretical (from last week):
Theoretical (from last week):







Key Concepts:










Back to Teamspeak Home