NJQRP "50MHZ" Frequency Counter

Based on the IK3OIL design this counter is a PIC 16F84 based device capable of having a prescaler added to its front end.

The original idea was to make a 2.4GHz capable counter. However, time constraints, component selection and other factors have led to this project being a simple 50MHz unit.

There are facilities on the board for a prescaler (included in the kit) but it was found that the chosen component did not start doing it's job until faced with a 250MHz or higher signal. Use of the original choice of either 2313 (1.5GHz) or 2314 (2.4GHz) scalers would require a change in the board design. Being that this kit was designed to be a "giveaway" at a club meeting it was decided that further expense would be spared. Update: 20070828 After further testing the counter has been found to operate as designed!! Operation has been confirmed on all frequencies upto and including 1300MHz. Testing at 2.4GHz was tried and results were gained but the tests were flawed due to faulty signal generation equipment.

The kit also has facilities for a serial port and is supplied with a MAX233 level shifter for future things such as frequency logging etc. It also has an input for an external clock source to allow most any known good standard (G7LTT GPS Disciplined Frequency Standard?) for greater frequency measurement.

Caveats & Gotchas

1) prescaler issues as described above.
2) ground pin missing on PIC
3) LCD display has word wrap
4) PIC source code not available as supplied
5) board has wrong chip size layout for RS232 level convertor
6) trimmer cap adjusts from bottom

Somehow, between the proof and the final cut the ground connection to the PIC was lost resulting in some very strange behavior. One must connect pin 5 of the PIC to a ground point. I used a resistor leg off-cut to connect pin 5 of the PIC's socket to the nearest ground point. I chose not to scratch away the etch resist but one could just as easily do that too. If not grounded the PIC will only operate when the VHF switch is selected thus providing an electrical ground to the device.

It has been found that the 16x2 LCD display has a word wrapping issue. This is NOT normal for a display of this type. They should display only the characters that will fit onto the screen and drop the rest. It is thought that there is an extra character such as a space which is forcing the wrap to occur. Again, this is a chosen component "feature". Seeing the counter as a platform for some experimentation, a 2 line display was selected. Update: 20070828 The word wrap issue has been tracked down to the firmware. The firmware was originaly designed for a 2x8 display. The word wrap is a function of the firmware's display instructions. New firmware is being designed to correct this and will be available shortly (see links below).

The sourcecode for the supplied PIC is not available. IK3OIL has declined to make it available to us. We do however have other source code that would allow one to create a new firmware. Please see the links below. Update: 20070828 New firmware is available in both compiled .HEX format as well as .ASM format. The files are available from the links below.

The component layout for the RS232 level shifter has been incorrectly sized. It is suggested that one soldered the left side of the chip to the left side pads and uses resistor cutoffs to bridge between the right side of the chip and the right side pads.

The selected trimer cap adjusts from the bottom. This is somewhat difficult in that there is no hole in the PCB to allow for this. Update: 20070828 CAREFULLY drill a 1/8th inch hole between the pads for the trimmer cap. This will allow access to the cap for frequency alignment.


Pretty pictures

My counter connected to my Frequency Standard as first built. Pretty close to being on frequency!

My counter when calibrated against the Standard.

The link I made for the missing ground. Click image to enlarge.

NU3E's counter