OCXO? Oh, You mean OCXO.

Well

Following my previous deliberations over how to align a frequency counter, I ordered a 10MHz High Stability Crystal Oscillator from China via ebay. I have decided to use it to make a frequency reference for my shack.

All I have done is box the device with a suitable voltage regulator and a power switch & indicator:

So, as a bit of a sanity check, I tried zero beating the signal from this oscillator with my Yaesu rig:

that all looked pretty damn good to me.

So all that remained to be done was to calibrate the frequency counter:

What do you recon? Good egh?

PIC 'n' Choose how to Count

Well, I've been fiddling with some PICs today, most notably the 16F628A. I bought the chip and a nice LCD display on ebay quite some time ago with the intention of building a good quality bench frequency counter.

So, to cut a long story short, another project is born. I've thrown together an initial prototype, but as is normally the case, this has raised a few questions.

Firstly, how do I align the PIC clock to be exactly accurate? With a frequency counter perhaps? Recursion - see recursion.

Then once I've done that how do I calibrate the counter? Here I would need one of those oven baked OXO thingumabobs. There's quite a few of them on ebay but they all seem to be in China; not sure I want to wait for a delivery from China.

So threes the basic board; there isn't much to it except a small RF amplifier and the PSU for the electronics, oh and the PIC itself of course.

The first thing I realised when I started to do this was that my PIC Programmer was in fact obsolete! Can you believe it; it wouldn't program these new fangled chips, so I trundled to a new local Maplin store and came away with a PIC programmer in kit form:

That took just over an hour to build. It also includes some I/O circuitry and other bits and bobs so it should come in handy in the future. The serial protocol doesn't seem to be compatible with MPASM from Microchip; so I'm having to use a non IDE which is a pain.

The display that I bought simply ages ago looks really neat when its being used:

It is, however, close to impossible to take pictures of.

I have a feeling that this project is going to roll and roll...

The following pictures are here for two main reasons, firstly because I can, and secondly for my mum:

This is Chopsey cat asleep on my new suitcase:

 This is the Geddy cat asleep in my left arm:

And  this is Chopsey cat again, but this time he is stupidly chasing his tail (he does this lots):

Fun,. egh?

It's running Hot Hot Hot!

I've been fiddling some more with my BITX:

http://g0mgx.blogspot.com/2011/01/boxing-bitx.html

and more specifically with the PSU that's built into it. I was using a 9-0-9 transformer, taking the 18V into the BITX. Once rectified and smoothed this was generating a basic DC supply of about 24V. Now, the 24V is a good value to feed directly to the IRF510 PA to get a good output, however, when I passed the 24V into the LM317 to regulate the value down to 12-13V the regulator was running really hot.

Thinking about the maths; then we have 24v in - 13v out = 11 * 0.8 (800mA) = 8.8 watts power dispassion! That's loads for a little LM317 to cope with even with the heatsync I've given it. I've been looking around for some time to try and find a suitable multi-voltage transformer such that I can have separate AC windings for the 13V regulated board supply and the PA. I found this on e-bay:

It was only the transformer I found on ebay, but I've built myself a power pack with a cabled 11.5V, +21V and 0V to the BITX. I'm about to drill the BITX case to take a 4-pin DIN socket and plug to act as the power pack DC connector so later today I'll be BITXing...

I'll keep you posted, but should be nice and cool!

Good egh?

How Noisy is this?

Well, there have been a few developments here of recent; firstly the FT-920 has gone and been replaced by an ICOM IC-756 PRO II. The radio is currently sat permanently WSPRing on 6m.

But getting back to Homebrew, I have been trying to make a noise source for HF; the idea being that I can use this for testing filter bandwidth. I stumbled upon the following schematic which is from the RSGB publication Hombrew Cookbook - strangely my wife looked at the receipt and wondered why I wanted what she thought was a recipe book for making wine!

 

And this is my dead bug style build of the circuit:

The observant amongst you will spot that I have a reverse biased zener diode in there in place of the transistor that has been substituted for same in the schematic. I plan to try a few different diodes and transistor junctions to see what difference it makes. I'm certainly not seeing quite what I expected from this circuit. This is what the spectrum analyser makes of the circuit output (the large spike in the middle is 0 MHz):

 and if I move the 0MHz point left and increase the scan width to 2MHz per division:

So I wasn't expecting to see such a dramatic decrease in amplitude as the frequency increased. I cetainly wasn't expecting the spectrum analyser to show the inverse image from 0MHz and below!

So, I'm planning to try a few different zener diodes to see what the difference is; the theory is that there should be a good amount of white noise generated when a zener is reverse biased to the point where it reaches its reverse breakdown voltage.

Weird results, egh?