Re: Buying custom crystals...
OUCH. No wonder why it was railing, it thought the frequency was ZERO because it couldn't detect any pulses from the VCO! I was hoping I could tune the VCO's amplifier to hit CMOS VIL/VIH but this is not the case. So I'll need to amplify it further before I can feed it into the divider
I then decided to write in a tiny frequency counter into the microcontroller because I grew tired of using my frequency counter... now I don't need to lug the counter around to test. It actually worked quite well despite having a oddball timing pulse source (I have a 28 1/8 Hz heartbeat clock, don't ask why...), plus I did not gate, so I lose pulses during computation. It only has 4-digit resolution too. Alas such is life, compromises. Still good enough to know if I'm in the ballpark, a quick go/no-go alive or dead test: at least tell if the input is reaching VIL/VIH thresholds.
Hmm...might have to cheese this with a Schmitt trigger inverter/buffer IC or something, consistently biasing a BJT seems annoying. Except I may need to shift the zero point... And I really need to design something that would work with 48MHz not just 8MHz that it is right now. So no op amps, or none of the cheap op amps I have in my collection. My 17MHz op amps are not fast enough
More analog trouble. Then again perhaps a voltage controlled device is what I need, else I would need a differential pair such as in the op amp... hmm... think think think. Such is electronic design 
OUCH. No wonder why it was railing, it thought the frequency was ZERO because it couldn't detect any pulses from the VCO! I was hoping I could tune the VCO's amplifier to hit CMOS VIL/VIH but this is not the case. So I'll need to amplify it further before I can feed it into the divider
I then decided to write in a tiny frequency counter into the microcontroller because I grew tired of using my frequency counter... now I don't need to lug the counter around to test. It actually worked quite well despite having a oddball timing pulse source (I have a 28 1/8 Hz heartbeat clock, don't ask why...), plus I did not gate, so I lose pulses during computation. It only has 4-digit resolution too. Alas such is life, compromises. Still good enough to know if I'm in the ballpark, a quick go/no-go alive or dead test: at least tell if the input is reaching VIL/VIH thresholds.
Hmm...might have to cheese this with a Schmitt trigger inverter/buffer IC or something, consistently biasing a BJT seems annoying. Except I may need to shift the zero point... And I really need to design something that would work with 48MHz not just 8MHz that it is right now. So no op amps, or none of the cheap op amps I have in my collection. My 17MHz op amps are not fast enough
More analog trouble.
Then again perhaps a voltage controlled device is what I need, else I would need a differential pair such as in the op amp... hmm... think think think. Such is electronic design
Comment