Category Archives: Uncategorized

Time Based Reference in VK3PKT’s Shack

So over the last 12 months the Twisted Shack of Mark VK3PKT has been getting a lot more test gear. Some of it was from the hamfest, some from people that I know and some from eBay.


I first aquired a 550MHz Freq Counter this is connected to a 10MHz GPSDO timebase to lock it to a known reference. It is then connected to a 10 MHz-520 MHz Marconi AM/FM Signal Generator to give an accurate display of the frequency of the RF Sig Gen.

1PTAX0001Marconi RF SIgnal Generator and Philips Freq Counter 


So I have a lot of my gear, 3 Frequency counters and a Time Interval Analyzer hooked up to a GPS Disciplined Oscillator. This uses GPS satellites to create a very accurate 10 MHz signal to use as a reference for the test equipment.  I also have a Rhode and Schwarz Frequency Standard that outputs 5 MHz 1 MHZ and 100 KHz. I use an oscilloscope connected to both the 5 MHz output of the R & S Standard and the GPSDO to get the FRQ standard on spec.


This scope is dedicated to the job of using lissajous figures to monitor the GPSDO and both the R & S as well as my OCXO Ovenized Crystal that runs off 12 volts and has a small SLA battery to keep it running while taking it from the shack to the car then to a mates shack to calibrate gear. I first adjust the transfer standard against the GPSDO using a Scope and then take it to the required location as needed.

transfer2 transferstandard1 transerfreqFrequency Counter showing Frequency of the transfer standard 

Now when it come to calibrating counters and reference standards I start by using the GPSDO as the Master Standard, this is the one that for my shack I trust the most to be accurate. That is what I feed it into a lot of my test gear as a reference. To check the 5MHz R & S I could just use lissajous figures but I have a bit of gear I picked up dirt cheap on ebay. It cost me 160 bucks delivered and new was over US $8000. It is a Yokogawa TA320 Time interval Analyzer. It measures periods down to picoseconds. Here is a histogram of the the 5 MHz signal calibrated against the GPS 10 MHz signal.


As you can see the range of the window is only 2 nanoseconds, from -1 to +1. The average signal is 199.97958198 over 99.999999 Million samples. The gate time is set against the 10 MHz reference. Apart from setting standards what would you use such a beast for ? Well the manufacturer gives a few ideas with one being connecting it to a hall effect sensor counting teeth on a flywheel of a drag car engine it is fast enough to see a stretched pulse at full throttle if the engine misses a bit. It is also used a lot to measure jitter in optical drives for CD’s and DVD’s in a musical recording studio making master discs.

TA320 with Freq Counter permanently connected to it

PTAX0001Here is a photo of the Rohde & Schwarz Freq Standard and GPSDO mounted in 19 inch Rack, there are four 10 MHz outputs on the back of the GPS that break off to test gear in the lab. One comes to the a hole that was in the front of the R & S that had a broken neon lamp in it and was just the right size for a BNC connection to break out the 10 MHz from the back of the GPSDO to the front panel to make it easy to get to it.


DIY RF Adaptor Set

So I was looking at commercial RF adapter kits and the prices shocked me so I decided to make up my own. The commercial ones I saw used a weird connector to join the adapters to each other. I decided I would use female BNC as the interconnect that way i could use male to male adapters to make short connectors or BNC patch leads for longer cables. Here is a commercial kit from element 14 it is pretty limited and is more than US$200.42421688

I put together a much larger kit with two of all the adapters as well as to 50 dummy loads just small ones to use to match the load when connecting to a scope to match the impedance, you use a BNC Tee piece. Anyway here is my kit… It has double of each connector and should cover all my needs it cost me about 120 bucks Australian to put it together.


Dust PC from Hell

So I rebuilt a friend of my sons PC. It was a mess it had water cooling and the radiator was hanging lose. It was overheating from all the dust and when he brought it over in the car the radiator ripped the CPU out and bent dozens of pins.


It was one of the worst PC’s I had seen in a long time for dust and if it had gone to a shop they would have thrown the CPU in the bin and it was an AMD FX 8 core worth 300 bucks

Nearly an hour under a stereo microscope with tweezers that are so fine they are like needles was spent getting all the pins straight.

I pulled it all apart except for the motherboard, blew all the dust out, mounted the radiator to the top and put a HDD cage in as the hard drives were just hanging by the power cables.



It took Angus and I about four and a half hours to get it from the first photo to the finished product in the last two photos, one of the back which has a case cover over the cables and one of the front side which is missing the cover.

Some people might think it is to much to repair a PC like this and rebuild it but I actually enjoy this type of work, I just wish I could actually do this as a full time job but my health wont let me

IMG_20160114_222123 IMG_20160114_222142


Okay I was a geek at school, .. Nothing has changed… Just ask my wife
I run Linux
I am a licensed Ham Radio operator
I Own 4 Oscilloscopes
I also own 8 calculators including 3 graphing
3 slide rules
I have a glass world globe on my desk
4 Ham Radios in my shack
3 Radios in the car
2 Handheld ham radios
5 Multimeters
My phone and tablet run custom android roms
My bookshelf in my man cave has the following books…
Machine Tool Operation Book 1
Machine Tool Operation Book 2
Taps and tapping by sutton tools
The Australian Amateur lapidary handbook
Meters for measuring water
RCA receiving tube manual
108 uses for an oscilloscope
Audio Handbook No2 “Feedback”
Phillips Valve data book
Electronics Made simple
Oscilloscope equipment
101 ways to use your signal generator
99 ways to use your oscilloscope
How to use grid dip oscillators
101 uses for your Vacuum tube volt meter
Transistor radios Circuits and servicing
The casio PB100 Computer manual
RCA Tube Manual
WIA Handbook Vol 1
Summary of lectures Automatic control systems
Faber castell slide rule manual
Hemi 40RK Sliderule manual
Icom ICT90a Manual
Transistor Transmitters for amateurs
Radio Data Reference book
Program Design
The boys book of crystal sets
ARRL radio Amateur handbook
Electronics Principles Integrated and discrete
GIMP Graphics Package User Manual (Linux)
Tech topics radio handbook
Radiotron designers handbook
High quality sound reproduction with valves
Temperature measurement
Novice Operators Theory Handbook
Tektronics Scope 314 Service Manual
1972 Melways (Street Directory of Melbourne) from my birthyear
The ARRL handbook 1995 (Theory)
Sourcebook of electronic circuits
Yaesu FT707 Service Manual
Ford Falcon EA-EF Factory Manuals (We own 2 falcons)
Running Linux
Linux Sys Admin
HEMA World Atlas
Oh and I own two domains, and
Geek and proud

Jaycar MP3086 30 Volt PSU Teardown and testing

Old MP3086

Okay so I bought a 30VDC Variable 3 amp PSU from Jaycar a few weeks ago. It is model MP3086 with CV and CC modes. I was expecting the one shown above with 3 pots, one for amps, and coarse and fine ones for volts.. Instead I got a new improved model, unless you like the coarse/fine pots!

New MP3086
So of course with anything new you need to follow Dave from EEVBlog’s advice.. Don’t turn it on tear it apart so that’s what i did..
Cover off, transformer looks nice doesn’t it ?
Okay so first impressions from the photo above.. Nice secure soldering on the mains connections, shake proof washers on the earth wire and extra covering on the mains feed, see that black rectangle on the right centre ? Here it is closer up…
Switch Close-up
It is a real clunking power switch on the mains, so no standby current usage, Yay a thumbs up..
IEC socket with integrated fuse holder, nice one !
Shake proof washer a point for Jaycar
Spring washers on transformer mounts with intake vent for fan, there is also vents on the sides with exhaust at the rear of the unit
Terminal block for permanent connections say a bench Multimeter and banana sockets

Okay now for the close-up internal shots, there are five boards

  • Rectifier board Power transistor board with Relays to switch taps from transformer to transistors
  • LCD Display board
  • Keyboard Board for buttons
  • CV/CC Control board
  • Connection board for front panel outputs with current shunt for amps readings

Rectifier Board with room for two more transistors

Is this a real Japanese Rubycon or a copy ? 
The Main Power board looks like it has a real Rubycon capacitor, the screen printing looks good and it does not look like a rip off

LCD Control Board all SMD, the bodge wires on the right run to the back light, keyboard board bottom left with ten turn pot for volts selection underneath
The Voltage and Current control board, there is nothing on the other side apart from a few through hole caps and a voltage regulator on a small heat-sink, notice the flux residue on the through hole hand soldered joints, not good how hard is it to clean a a board ? Here are some close ups of the cruft left behind after soldering. But hey the Ningbo QJE company gave us their phone number if you want to call and complain to them about it :p

NXP Chips a counter IC and a logic gate, some more quality parts to go with the Rubycon cap…

Nice SMD soldering, re-flowed not hand done ! This is one of the trimmers to calibrate it all
Nice thick bus bars to the connections on the front panel with shake proof washers as well, excellent job.
Thermal switch for the fan, wired in series with fan it is rated at 5 amps and 50 C cut in temperature but cuts in before that, 
Okay now for some testing…. 
The test set-up consisted of the following.
  • RS232 connected DMM logging temperature
  • Galaxy Note 2 Smartphone for dB readings (Not real accurate but okay) 
  • Incandescent 12V globe as resistive load
  • Digitech QC1932 25MHz DSO
  • Supply set at 6.9 Volts, this just switched to the second tap (6.8V) so max heat from transistors

12 Volt Light globe pulling 1.33 amps at 6.9 volts, this tested at about half of the maximum amps and the second tap on the transformer had just switched in at 6.8 volts so I was dissipating the most heat for this amperage in to the heat-sink. 
The test set-up, I connected the scope probe to the terminal block on the front directly, I just hadn’t done it here yet
Below is the capture of power on unloaded, there is a small HF ripple at power on then ramp up to full voltage (5VDC) with no overshoot, there is a small amount of ringing when it reaches the set voltage, from power on to stable voltage is about 10 milliseconds.
This is a close up of the spike at power up unloaded, about .5-.75 volts lasting 206 microseconds
Here is a capture of a loaded supply at 7VDC from turn on to full output is 76 milliseconds with no overshoot to speak of.
1.6mV ripple at 7 volts under loaded conditions
dB levels with phone acting as meter with smart tools app. The first bump is me pressing the power switch, the second is the mechanical temp switch clicking in, then you can see the fan ramp up to 69dB. the phone was sitting 10cm behind the PSU with the microphone sideways to the fan, the power switch click is quieter as it is at the front while the temp switch was closer to the rear. The ripples in the graph is just background noise in the shack.
If you look between the cursor marks on the timebase below you can see two little ripples lasting about a second, this was when the fan switched in but was brief and low in amplitude. It may have just been induced noise from the arcing in the mechanical temperature switch for the fan as its contacts closed.
Here is the temperature run at 7VDC 1.33 amps from cold (I did this test first after leaving it off overnight) Starts at about 26c the shack temp, ramps up to 43.9c in about 2 minutes then settles to around 34c and stays there. I ran it longer then this graph loaded and the temp was stable and didn’t run away at all which it shouldn’t pulling 9 watts from a 90 watt supply.

 All in all for the price I am very happy with it, well built steel case, the front panel is plastic but has steel rails running from it to the back panel for support.

So what don’t I like ? Well not much the only things are as follows.

  • Earth Connection is in the centre, so I cant plug a standard spaced dual banana plug adaptor in to the – and + rails, I may re arrange the banana sockets and move the earth to one side.
  • You need to short the rails to set the required amperage for CC mode, I am going to put a NO momentary switch on the front so i don’t need a lead to do it
  • Switching on CC mode is a bit hit and miss, I dont know if it is key bounce or a slow micro but you have to press the CC button a few times before it locks in to CC mode, it is a soft switch
Apart from that i really like it and it actually goes to 31.5 Volts
MP3086 Regulated lab Power Supply 30VDC at 3 AMPS
List price $199 AUD