Guru
Joined: 02/02/2017
Location: AustraliaPosts: 1424
| Posted: 12:15pm 03 Nov 2017 |
Part 16: BYO clock for the EG8010
As a result of my past trials and testing, I have a few defunct EGS002 boards.
One of which still has a functional EG8010. "you'll do", I thought.
I have a need to achieve sync of the EGS002 output with our local street power.
This is to aid in the event when the house switches over from using street power to using inverter power. At present this event occurs at any time in the 50Hz cycle.
Sometimes bad things happen. I plan to build a mains sync subsystem, creating a clock for the EGS002 such that it remains locked with mains power.
After removing the 12MHz crystal, I attached two wires to the pads where it was located pin 10 and 11, with a 220nF ceramic cap in series with each wire as AC couplers. I set the function generator to output 12 Mhz with a low level of 0V, and high of 5V. I attach the fungen ground to EGS002 ground. The fungen output can be put on either pin 10 or 11.
Apply power and it works.
Sometimes it does not start up, maybe due to bounce when I manualy attach power.
In any case beware of letting it sit powered up but not running. I saw the EG8010 chip get so hot as to burn my finger once. That was when I put the fungen ground on pin 10 and fungen output on pin 11 in an unsuccessful attempt. It needs to oscillate.
Anyway, I’ve seen it run down to 2Mhz, up to 20Mhz.
Time to have a look at tunable ceramic resonator circuits. Crystal oscilators can’t be tuned over a wide enough range for my application. Typical tune ranges of +/-100 parts per million range is far less than the mains freq deviation present here in Melbourne.
I’ve seen it down to 49.85 and up to 50.15 on a daily basis.
100 ppm is equivalent to 49.995 to 50.005 Hz range.
wronger than a phone book full of wrong phone numbers