Pixie Perfection

As has been noted in several posts, the stock Pixie 3-pole filter is hopelessly inadequate to suppress harmonics.  A typical Pixie will have a 2nd Harmonic that is only about -21 dB down from the carrier.  An external 5-pole filter did not improve things well enough, and ultimately a 7-pole external filter was required.  In both those cases, the original 3-pole filter remained in place.

However, what if the original Pixie filter was replaced with a 5-pole design?  Would that be enough?  A friend of mine, Willy W1LY, set out to discover just that.  He pulled the original axial leaded inductor (L2)  and replaced it with a pair of T37-2 toroids.  He then paralleled additional caps on the input and output side (C5/C6), and finally added a new cap between the two toroids. He was able to fit all of that in the space of the original filter plus a bit of open area next to the BNC connector.  In addition, Q2, the S8050 output transistor, was replaced with a P2N2222A.

Elsie 1 dB ripple 5-pole filter. Note, it shows 2 caps between the toroids — only one cap is required

  • C1/C3 input caps were set to 910 pF (standard value — Willy used 940)
  • C2 cap was set to 1300 pF (standard value — Willy used 1270)
  • L1/L2 were set to 1.2 uH (20 turns #26 on T37-6 core)

The results were pretty much ideal.  The 2nd harmonic was suppressed by 63 dB (other harmonics were even lower).  The fundamental power was increased to 25.2 dBm, or 330 mW.  (Spectrum Analyzer is fed through a 40 dB external attenuator)

Given the current 5-pole LPF filter values are producing a 2nd harmonic rejection that is about 20 dB more than required, there is some room to improve the ripple response of the filter, which would probably boost the output power to 450 mW or higher.  But 330 mW is more power than any stock Pixie has ever produced (usually they are around 200-250 mW out at 13.8 volts input).  Another experiment would be to change the toroids out for 1.2 uH axial inductors.

Posted in Pixie, Projects, QRP | Leave a comment

What’s wrong with the serial port on the Raspberry Pi 3?

The Raspberry Pi 3 added all sorts of features to the Pi, including a much faster speed adjustable CPU, built-in WiFi, and Bluetooth interfaces.

Some of the hardware interfaces on the Pi continue to have limited functionality.  Such is the situation with the RPi 3’s serial port.  The primary serial port, which has a fully adjustable and stable baud rate generator, was dedicated to the Bluetooth interface.  That meant that the “User” serial I/O port was assigned to the secondary serial port.  This is a major issue as that serial port lacks an independent baud rate generator, but uses the system clock instead.  The problem with that is the system clock is dynamically changing all the time based on CPU load (to provide bursts of higher speed clock rates without causing the system to overheat or draw too much power on a long-term basis).

Because of the variable system clock rate, the user serial I/O port is almost useless.  For example, you can’t hook it up to a GPS Hat and expect it to work since the baud rate changes unpredictably.

There are two workarounds:

  1. Run at the slowest clock rate all the time, crippling the RPi3
  2. Disable the Bluetooth interface and assign it to the user serial port

More details here:  https://forums.adafruit.com/viewtopic.php?f=50&t=94677

And here:  https://openenergymonitor.org/forum-archive/node/12311.html

And here:  http://raspberrypi.stackexchange.com/questions/45570/how-do-i-make-serial-work-on-the-raspberry-pi3



Posted in Raspberry Pi | Leave a comment

P2N2222A for the Pixie

On Semiconductor has a variant of the old 2N2222A in a plastic case called the P2N2222A and characterizes it as a small signal RF amplifier.  I picked up a few samples and ran three of them through the Pixie.  In all cases the RF output had increased to about 575 mW on the meter, with the SA reading 525 mW.  The second harmonic was well below FCC limits at -57 dBc.  That’s about 1.7 watts total input power on transmit, or about 1.15 watts dissipated in the transistor and filters.  Gotta keep that key-down period short!


Posted in Pixie, Projects, QRP | 1 Comment

7-Pole Filter + Pixie

A friend of mine, Paul K1YBE, gave me a QRP Labs 40-meter Low Pass Filter Kit.  The kit came with three T37-6 cores, some #28 wire, 4 caps, a tiny PCB, and a couple of headers. Normally used with one of their WSPR Transmitter Kits.

The filter was easy to build, but I found the inductors needed to have a turn removed to yield their desired value.  This equated to 20 turns of #26 wire for L1 and L3 (1.37 uH Q17), and 23 turns for L2 (1.81 uH Q15).  The caps had a measured tolerance of +/- 3.5%. The 270 pF caps had a Q of 30, and the 680 pF caps had a Q of 260 (all Q values at 100 KHz).

Assembly was straight forward, and I used my Quad-Hands to help keep things steady while soldering – especially the flying leads I was going to use to mount the filter in the box.

I mounted the assembled PCB into a Pamona 2391 BNC box.  This is a great solution because the box will connect directly to the Pixie output using a Male BNC, and provides a filtered output on the female BNC.

The filter response was swept with my spectrum analyzer.  Being a 7-pole design, the loss was much less at 7.040 MHz, only about 0.3 dB down.  At 14 MHz the response was down about 36 dB, and 39 dB at 21 MHz.  In truth, had I not removed a turn for the two inductors, the 3 dB point (8.76 MHz) would have shifted a bit lower, probably providing a tad more rejection at 14 MHz.  Given the tiny box it was easy to hook the filter to the SA.

The real test was hooking the filter to the output of the Pixie.

Looking at the output on the SA was pretty gratifying:

The 2nd harmonic is down 59 dB and the 3rd is down 69 dB from the carrier.  The SA shows  26 dBm output (400 mW).  The power meter shows 450 mW  (these tests were run using a 2N2219 for Q2).

Posted in Pixie, Projects, QRP | Leave a comment

Better Q == Better Response

Yesterday I replaced the inductors in my original 40-meter LPS with larger cores that wound up having about twice the Q.  This morning I replaced the ceramic caps with silver mica caps.  The old caps had a Q around 80.  The new caps have a Q > 1000.  As one might expect this had an impact on the insertion loss of the filter, lowering it by almost a full db from -1.97 down to -1.03 dB.   The tighter cap tolerances also resulted in a cutoff frequency of 7.26 MHz, probably a bit too close to the operating frequency.  Rejection at the 2nd harmonic was -39.2, and at the 3rd harmonic was -35.0 dB.

Actual transmit performance using my original S8050 transistor (which had 2nd harmonic suppression of -10 dB) had a 2nd harmonic suppression of -51.9 dB.

Ceramic Caps Silver Mica Caps
Meter Power 270 mW 240 mW
7.040 MHz -16.7 dBm -19.4 dBm
14.08 MHz -61.6 dBm -71.3 dBm
21.12 MHz <-80.0 dBm < -80 dBm

Note that the measured meter output power was 240 mW, which is down about 30 mW from the ceramic cap solution.  The SA indicated the fundamental was at +20.6 dBm or 115 mW).  Given the lower insertion loss, I suspect this is because the filter with the SM caps and T50-2 toroids is providing a sub-optimal impedance match to the output transistor.

Putting the 2N2219 transistor in the circuit boosted the output power up to almost 280 mW by the meter, with a 120 mW fundamental by the SA., and a second harmonic suppression of -50.6 dB.

Posted in Projects, QRP | Leave a comment

Does Size Matter?

There’s a joke in there somewhere…

Yes, size does matter when it comes to Toroid diameter (which also has an impact on winding geometry), maximum wire size, etc.  All things which impact the Q of the circuit.

I went back to my original S8050 Transistor, and swapped out my T37-2 inductors for T50-2 inductors wound with 15 turns of #20 wire (13″ of wire, measuring 1.2 uH).  This allowed the winding to be single layer and evenly spaced (the smaller T37 cores had to go to two layers and were really jammed in there).

I had seen Q readings at 100KHz of 20 for the T37-6, 25 for the T37-2.  The larger T50-2 inductors had a Q of 45.  The higher the Q, the lower the insertion loss, so larger cores with heavier gauge wire are better.  That’s the same conclusion that is reached in the Micrometals Q Curve App Note (I’m using Micrometals cores).  The filter response curve was very similar but at 7.040 Mhz the loss was now -1.97 dB, about a 0.1 dB improvement.

With the filter using the T50-2 cores the spurious response output looks like this (Meter reading just about 270 mW, 2nd harmonic down -45 dB, fundamental at 220 mW).

T37-2 17 turns #20 T50-2 15 turns #20
Meter Power 250 mW 270 mW
7.040 MHz -16.6 dBm -16.7 dBm
14.08 MHz -63.0 dBm -61.6 dBm
21.12 MHz <-80.0 dBm < -80 dBm

Proving again that basic physics never gets old, here is a great reference from Boonton Radio Corporation, circa 1955, that describes a Q-Meter and how you can evaluate components.  (Thanks to John, WA1ABI, for this great article!)

Posted in Projects, QRP | Leave a comment

A New Transistor for the Pixie

Of course the Pixie final transistor, Q2 S8050, gets darned hot, and I’ve blown one out in the past just holding the key down more than a few seconds.  With my Pixie + Red Core Filter, I was seeing about 250 mW out on my QRP meter.

I decided to replace Q2 with an old 2N2219 NPN transistor (one of the few metal can devices I had in my part box).  Sure enough, that old device, circa 1975, had better performance than the S8050.  My output meter was now about 320 mW out (SA was reading +24.3 mW or 270 mW), and the 2nd harmonic was 50 dB down.  A new 4th harmonic showed up, but was also 50 dB down.

Playing around with the output transistor will likely yield some improvements.

When viewed from above, with the board rotated so the BNC connector would be on the right-hand side, Q2 will be flat-side toward the bottom edge of the board, and the pin-out is EBC from left to right, as shown below.

Posted in Gear, Pixie, Projects, QRP | Leave a comment

Passing Pixie!

After experimenting with T37-6 yellow core inductors, it was time to change to the T37-2 red core.  These have a lower operating frequency, up to 10 MHz, and it is possible that they might have a higher attenuation on 14 MHz and up because of that.  The toroid calculator indicated that 17 turns would produce a similar 1.2 uH coil.  That worked out to 12 inches of #20 wire, still more than a single layer can deal with, but I’ll put up with 4 turns of overlap.

It was immediately apparent that the filter response was different.  The loss at 7.040 MHz was lower and the attenuation at 14/21/28 was higher by a few dB.

Yellow 19 Turns Red 17 Turns
7.04 MHz -2.33 dB -2.07 dB
Knee -3 dB 7.84 MHz 8.00 MHz
14.08 MHz -35.4 dB -38.4 dB
21.12 MHz -29.0 dB -31.3 dB

When added to the output of my Pixie, which was producing lots of harmonics (14 MHz was only 10 dB down from the carrier), the red core filter was a substantial improvement compared to the yellow core.

The second harmonic has been reduced 46.4 dB below the carrier.  The Pixie is Passing the FCC limit of -43 dBc.

Yellow 19 Turns Red 17 Turns
Meter Power 230 mW 250 mW
7.040 MHz -21.6 dBm -16.6 dBm
14.08 MHz -60.9 dBm -63.0 dBm
21.12 MHz -77.1 dBm < -80 dBm

The Pixie has the Red Core Filter in series with its output.  The output of the filter is fed into an external 40 dB attenuator.  The attenuator output is fed into the input of the Spectrum Analyzer.  Thus the measured 7.040 MHz signal at -16.6 dBm is actually -16.6 dBm + 40 dB, or +23.4 dBm.  That is 219 mW.

As one might expect the output of the Pixie now looks like a real sine wave with the harmonics suppressed.  By measurement, the signal is 9.4 vpp, or 3.3 vrms.   That corresponds to 221 mW into 50 Ohms.


Posted in Gear, Pixie, Projects, QRP | Leave a comment

Pixie PLUS 5-Pole LPF

My Pixie produces about 500 mW of output power when driven by a 13.8 volt supply.  The total input power to the board is about 1.5 watts.  As I’ve mentioned the Pixie only suppresses the 2nd harmonic by about 10 dB with it’s internal filter.

Pixie running at 13.8 volts, producing 500 mW output, with second harmonic down 9.5 dB from carrier

When I added the 5-pole filter that I described in my last post to the Pixie minus one turn on each coil, so 19 turns each (keeping the original filter in place), the output power was reduced to 250 mW, but the 2nd harmonic reduction was improved to -40.7 dBc; almost passing!

Pixie running at 13.8 volts, with 5-pole filter on output, producing 250 mW output, with second harmonic down 40.7 dB from carrier; just 2.3 dB above FCC limit


Recall that other Pixies I have tested have 2nd harmonic suppression of about 20 dB.  That will probably allow the additional filter to meet FCC requirements.

I also removed another turn off of each inductor, but only observed a slight improvement in output power (it went up to 280 mW).  The 2nd harmonic suppression was negatively impacted as well, now only 37.5 dB below the carrier.

Posted in Pixie, Projects, QRP | Leave a comment

A 40 Meter LPF for the Pixie

As I mentioned the 40-meter Pixie QRP rig has pretty horrible harmonics, well above the maximum allowable FCC limit (-43 dBc).  One Pixie was about -9 dBc, and the other was about -20 dBc.

The output filter in the Pixie is a simple 3-pole PI network, clearly not up to snuff.  With judicial parts placement, an extra 5-pole filter could be installed, but it probably makes sense to install it in an external box (Like a Pomona 3231 or 2391).

Several club members and I embarked on experiments to build a prototype filter, with some different choices of -3 dB cutoff frequency, output ripple, etc.  We also discovered that while many of the online calculators produced very similar results, at least one of them does not.  Filters built using those values did not match up well with predicted results.

John King, WA1ABI, suggested using “Elsie” from Tonne Software.  Far more flexible than the online calculators, it produced output values that were consistent as long as the initial parameters were correctly selected.  The software can be downloaded here:


I used an online calculator that produced identical results to Elsie:


Input values were a cutoff frequency of 7.5 MHz, 50 Ohm impedance, 1 dB ripple, and a 5-pole Chebyshev filter.  That produced the C1/C2/C3 & L1/L2 values shown below:


C1 & C3 were fabricated from a 680 pF in parallel with a 220 pF, for a total of 900 pF (less than 1% low).

C2 was fabricated from three caps in parallel; 470 pF, 470 pF, and 330 pF, for a total of 1270 pF (less than 1% low).


My inductor was set to 1.2 uH, about 4% high, but that required 20 turns on a #6 core.

I had chosen T37 cores for their small size and had initially planned on winding them with #26 wire.  However I switched to #20 for mechanical reasons.  A T37 core can hold about 13 turns of #20 in a single layer, and 20 turns of #20 in a dual layer — perfect.  14″ of #20 magnet wire was enough to wind 20 turns with 1.5″ long pigtails.  It was tight, but things fit.

The final breadboard circuit looked like this:


The test setup looked like this, with the breadboard hooked up in series between the Tracking Generator output and Spectrum Analyzer input:


The measured response was:

  • 7.055 MHz -2.6 dB (Operating Frequency — loss is higher than expected)
  • 7.429 MHz -3.0 dB (Cutoff Frequency — a bit lower than expected)
  • 14.112 MHz -35.1 dB (2nd Harmonic)
  • 21.168 MHz -29.7 dB (3rd Harmonic)


Posted in Gear, Pixie, Projects, QRP | 1 Comment