Homebrew PiHPSDR

Having seen the Apache-labs version of the PIHPSDR I wanted to customise it to fill my needs, so I needed to assemble my own

All the needed information , with the software,  is at John Melton's github site https://github.com/g0orx/pihpsdr  The hardware shopping list includes. RaspberryPi 3, 7" Official Raspberry Pi LCD,  8 push buttons, 4 rotary encoders, case and power supply.

All the items were mounted in a 12x7x2" aluminium case obtained from Mouser, The display was held in place with plastic channel finishing strips from B&Q.

Front Panel View

Inner View

Initially I used a $30 rotary encoder that I got off ebay, but being designed for attaching to a motor it was a bit big to go in the box. I replaced it with an HP  HRPG-ASCA # 14F which cost $5 off ebay which fitted nicely into the box.

When first built, I used a 5V 3A regulator bolted to the Aluminium case. However this was dissipating around 8.3V at 2A and was too hot too touch. I tried a series of diodes, each with 0.6V drop on the regulator input, but that also got hot.  The regulator was replaced with a DC-DC switching converter from ebay which kept cool. The negative lead of the power supply, and the negative lead of the RPI were both grounded to the case to avoid negative lead voltage drop issues which caused RPI brownouts

I also included a USB soundcard in the box so i could listen to both audio channels when remote from the ANAN. The audio connections were extended to sockets mounted on the Aluminium box. Due to the close spacing of the USB ports on the RPI, and to allow plugging other items into them, the plastic case on the USB soundcard was removed.

The total cost of the homebrew PIHPSDR was around £120

AQRP VIA Power Source Improvement

Running the VIA from the rechargeable Eneloop cells from the VIA was tolerable for a while, but the batteries had to be extracted from the battery holders in the unit every time a recharge was needed.

The VIA documents mentions using a "2600mAH power bank". These provide a regulated 5V output from a USB connector and can be charged from a phone charger with microUSB plug.  Ebay had them for 99 pence (shipped) so a couple were bought.

Testing them to power the VIA externally proved they would work fine. The BIG issue was that  power bank box was too big to squeeze into the VIA so the power bank contents would have to be extracted from the package. This was done but it was quickly discovered the controller boards are destroyed if 5V from a charger is applied without the battery connected. The LIPO cell still worked. Time to look for a more robust charger for my growing LIPO cell collection

Adafruit had a Powerboost 500 charger/booster which looked to be well suited. It also had the advantage that the VIA could be powered at the same time as the cell was being charged. One was obtained for $15 and squeezed into VIA with the LIPO cell, as shown below.

Controller  shown top Left, just above the LIPO cylindrical cell

The system powers the VIA for well over an hour, which is very acceptable. A hole was made in the case to allow the charged and charging LEDs to be viewed with the case lid on.

Austin QRP Club VIA Finally Built

I had seen the specifications for this Austin QRP Club Vector Impedance Analyser kit when it was first announced. It covered 1 to 150MHz or 8kHz to 1.17MHz (good for LF) and offered a range of different format results display on it's TFT Display; no need for a computer to drive this one!  I bought the kit from K5BCQ's web page about a year ago and recently decided it was time to build it. The microprocessor board is prebuilt, the TFT display is prebuilt, the only board which needed building is the RF board which also connects the other two boards together. The RF board took about 2 hours to build

The recommended enclosure was obtained; not cheap at around 20 pounds, but makes for a neat unit. I used an SMA connector for the antenna port.

Not surprisingly the software had been updated since I bought the kit, but it didnt take long to do the upgrade following the instructions provided.

Initially the unit was powered by six Eneloop AA cells mounted in the supplied battery holders connected via the regulator board supplied with the kit providing regulated 5V

The flash screen came up immediately and following the instructions in the manual a 2 MHz to 30 MHz open/short/load calibration was done. My Alpha-delta DXCC antenna was connected and the VSWR, Return Loss and even a smith chart plot were obtained

VSWR Plot

Return Loss

Smith Chart

It is possible to export the data to a computer and process it in any way you want. 

 Another useful display is the Alphanumeric display which shows the parameters as numbers. One use of this is to measure capacitors and inductors at RF Frequencies 

Alpha Display

Very impressed with the unit, especially as it cost under 100 pounds. Cannot wait until the transmission measurement add on becomes available

Triband 6/4/2m Amplifier finally finished

Last Summer, browsing ebay, I found a Kalmus 172F wideband amplifier. "Kalmus broad band vhf power amp model 172F High gain 200 watts rf output <1w @ 50-100mhz <4w @ 144mHz, on the tin it does say 80-160Mhz but it does cover 50-70MHz, supply required 28 volts dc 14 amps" Looked interesting so I bought one

In the last few weeks finally got around to using it. I wanted it to be smaller than it's bigger 400W brothers so i managed to squeeze it into a 12x7x3" box. No room for low pass filters so they have to be plugged in externally. A PIC oversees the amplifier and displays status on a front panel 20x4 LCD.

The kalmus amplifier already has a bridged T input attenuator, so I increased it by 3dB so I can drive it with the 7W from the Anglian / Nacton XV on 4/2m and 8W from the K3 on 6m. On all bands it produces over 200W

IFFY STM L7805CV devices from Farnell

This week I started work on my my triband 28V 4/6/2m amplifier using a Kalmus module I got off ebay. It needs a PIC controller which runs off 5V so I ordered twelve STM made L7805CV from Farnell which should give 5V output with up to 35V input.

 I connected up the PCB and LCD and turned it on. The LCD got very bright and the pic refused to work. Remembering the addage "99% of all problems are power supply related" I measured the voltage coming out of the L7805. It was 7.88V!  Replaced the regulator with another one from the same batch. 7.90V. As an experiment I ran the regulator off 13.8V. It gave 5.05V which is fine.
Looking in the component collection i found an old STM L7805. Soldered that in and it gave 5V with both 13.8V and 28V input. The only difference(apart from the date code) between the old and new regulators was that the new one said "STM CHN" the old one said just "STM". Enough said!

I contacted Farnell, who apologised and were persuaded to replace the iffy regulators with the Fairchild LM7805CT. These have now arrived and work fine off 28V!

Moral: Measure the output of regulators before applying to circuits!

LF season approaching

As the winter is approaching it was time to think LF again. When I bought the main parts for the MF amplifier from Dave G0MRF at the HF convention I also ordered one of his preamp kits. I managed to squeeze it into a small pomona box, time to try it out

Initially I connected it to my Alpha Delta DX-EE and ANAN-10 and left it receiving WSPR using K1JT's new1.7.0RC2 WSJT-X software. It seemed to do well:

Time	Call	dB	Loc	Power	km
18:22	G4FTC	-21	IO91pi	2	137
18:36	PA0A	-24	JO33de	1	591
18:38	LA8AV	-28	JO59cs	5	1151
19:56	G3WCE	-29	JO02pt	1	255
23:52	F1AFJ	-27	JN06ht	1	625
00:46	DJ0ABR	-29	JN68nt	0.2	1148
00:54	G8HUH	-19	IO81mg	1	103
01:16	PA3ABK/2	-25	JO21it	0.5	481

The preamp reduced the LOS droitwich 198kHz on my system.

During the evening  I wondered what was on 136kHz. So I dug out my old G0MRF 136khz preamp and let it run for a short period, while, again connected it to my Alpha Delta DX-EE and ANAN-10 receiving WSPR decoded 2 stations

Time	Call	dB	Loc	Power	km
22:00	G8HUH	-20	IO81mg	0.1	103
22:04	2E0ILY	-12	IO82qv	0.01	91

FAKE AD4351 in AD4351 Signal Generator?

Having got the signal generator from ebay seller ayanhu81 running it was time to measure Phase noise. I decided to start at the low frequency end, so 42MHz was chosen. The controller was set to this frequency and the lock light came on. However, looking on the spectrum analyser no signal was found. Winding out the span a signal was found at 2668MHz (42x64), very strange! Next 116MHz was tried. The controller was set to 116MHz and the lock light came on.  No signal was found at 116MHz but a signal was found at 3712MHz (116X32). I next tried 144,432, 1296 and 2320MHz and the output was produced on the expected frequency. Checking further the IC only worked correctly above 137MHz

As a check I used the AD435x software tool to tell me what the register values should be for 42 and 116MHz and modified the arduino software to load the hex values directly, but still got 2668 and 372MHz. Its not the code, its a problem with the chip

Looking ath the Analog Devices datasheet for the AD4351 it has a divide by 64 and divide by 32 circuit in the chip, which my IC apparently didnt have. I checked the markings on the chip and it said AD4351. I did notice from it's datasheet that the AD4350 does not have these dividers, behaving exactly as my AD4351 was doing currently. Obviously my board had an out of specification AD4351 or a mislabelled AD4350

At this point G4JNT made a post on the UK microwave reflector that he was experiencing exactly the same issue with exactly the same module from the same Chinese supplier

I have raised a "not as described" case with ebay against the seller. He was forced to refund the cost of the module!

Moral: You get what you pay for!

KR600RC Maintenance

A few weeks ago I uncovered the Rotator and the control box for my KR600RC Rotator. Connecting it together and powering it up resulted in a lot of whirring from the rotator and only clockwise movement. Removing the cover the reason for the whirring was obvious; the drive band had perished.  I searched the web for the a replacement there was no obvious source. However there was a hint from the VKL reflector that the rubber nitrile band was a #29. Web Research showed that in the UK this is called a BS029, which were available on EBAY costing  £5 for 10, so 10 were ordered

The reason for no counter clockwise was also investigated. The reason was because the NC contact on the clockwise switch, through which the counterclockwise switch gets its power, was open circuit. Note that the 2 switches are different; the clockwise is DPDT (6 pins) , the anticlockwise is DPST (4 pins). Researching a replacement switch, I found someone was selling refurbished ones in Australia Kur02.com.au, so it must be possible to take them apart!

The replacement bands arrived last week so it was time to start repairs. First the rubber feet were taken off. The cover was then removed and the front panel detached. The gear train then had to be dismantled, by removing the calibrated bezel screws to get to the screws holding the metal cover holding the gears in place!) to get the band in place. Note that the rubber band goes round 3 axles to maintain tension

Powering up the indicator then worked properly. The broken switch was then taken apart by releasing the 4 lugs. I noticed that the 2 contacts had slightly different profiles. The non contacting contact was then removed and bent into a shape matching the working contact. The switch was then reassembled and was now found to be performing properly. Job done!

My only concern is how robust the controller is for mountaintop operation; I can see building an all digital controller with LCD display coming up!

ADF4351 signal generator

Although I have a portable signal generator using the SI570 that can get to 200MHz I needed something that goes a bit higher. Looking on ebay I found some modules that cover 35 MHz to 4400MHz using the ADF4531 for 20GBP so I ordered one (auction 172152204149 by ayanhu81)

It took 4 weeks to arrive. The next question was how to control the module. An internet search revealed a design by F1CJN here  using an Arduino UNO and LCD shield with buttons from dfrobot.com, which I did not have. However I remembered in my collection I already had a similar LCD/Switch unit from adafruit This would need some code modification, time to remember how arduino's worked. It was not too hard as Adafruit provide the libraries and examples for this unit on their web site. The biggest issue was as my switch reading used SPI not the A/D converter of the original design it was much faster resulting in much keybouncing.  Some strategically placed delays stopped that issue.

I now have it working. Listening to the signal on 144MHz and 432MHz it sounds OK, if a little off frequency due to the cheap on board reference (6kHz low at 432.2MHz) but as the board has an external reference input that is not an issue. Next to take a look at the output spectrum, which is reputed not to be too good at lower frequency, then to put it in a box.

BEWARE OF THE GREEN BOARD VERSION! SEE UPDATED POSTING 

IF YOU WANT A PCB THAT WORKS TRY THE BLACK BOARD VERSION SEE LATEST POSTING

70MHz Trophy

Enthused about 4m from previous outings we decided to visit British (IO81KR) again. We were late starting by 30 minutes by eating cake and coffee before departing to the site!. There was mist/rain/drizzle for the whole contest. Again we used the K3S and the Anglian transverter to drive the I0JXX dualband amplifier. The antenna was the 5 ele Powabeam at 6m,

The only equipment issue was computer related. The K3S just has one USB connection to the computer (no audio leads). For the first 30 minutes on every speech peak the virtual serial port and the K3S soundcards would be removed from the computer and reappear on receive. After a while there was a warning that the device was disabled as it was unstable. Not having had this issue on previous 4m events I took a look at the USB lead. I had taken the first one from the bag; a thin black lead. Looking further I found a thicker lead with clear cover that showed a decent braid. Changing to that cable, the issue went away!
In a change to  normal, there were numerous stations QRV up north, in IO84 and 94. This activity allowed us to confirm that the site is good to the North, we dont normally work much up there as there is no activity! The loudest signal was from GM4ZUK/P in IO86. Although two stations were worked in GU, there was not even a hint of a GJ station or GD. Being called by GI4OPH in the last hour was a welcome surprise/ 98 QSOs were had with the best DX being PA4VHF in JO32 at 640km