NE555: ne555 oscillator frequency question

James,

Is this frequency "drift", cycle to cycle (jitter), device to device, or some other change?

1/2 Hz for a 50Hz signal is 1% accuracy. Getting accuracy that good or better with diodes, capacitor, timer, and resistor variances with their temperature coefficients and supply voltage will not be practical. It is easy enough to get rid of the diodes. Getting a very <<1% NPO/COG capacitor can be expensive. 

A crystal , crystal driver and a divider will be more accurate and less expensive than 0.1% caps and resistors. 

Hi Ron,

This is a cycle to cycle drift, not a device to device or other changes.

is it correct that the resistor/cap accuracy would only effect device to device drift? 

So if the resistor and cap is already on the board, the frequency should stay at some point at 50Hz or nearby 50Hz. And it should stay still, instead of drifting from 49.5 to 50.5Hz? 

Thansk.

James

Hi Ron,

Do you have any updates for this question?

Thanks.

James

James,

The main cause for cycle to cycle drift is AC voltage on VCC and control pin.

In the monostable section the data sheet says "The timing interval is, therefore,independent of the supply voltage, so long as the supply voltage is constant during the time interval."

In the a-stable section the data sheet says "As in the mono-stable circuit, charge and discharge times (and, therefore, the frequency and duty cycle) are independent of the supply voltage".

Hi Ron,

But the VCC/control PIN voltage doesn't seems to have some AC voltage change during all the time.

Could you please help to check if you have any ideas?

Thanks.

James,

Can you provide AC coupled VCC and control pin voltages so I can see the noise? 

Early next week, I can measure jitter in a clean setup to get a baseline.  

Hi Ron,

Thanks for the test.

Customer has a questions that if they only have 1% resistor and 10% capacitor, which would have huge impact on the output frequency.

Do we have other solution that we could keep the output frequency within 2% precision or even 1% precision under this limited 10% cap choice?

Thanks!

James,

I'll present a 1% or better solution tomorrow. I'm' not quite done yet.

James,

The basic idea is using a low cost 32,768Hz crystal supported by SN74LVC1GX04 then divide that by 328 using CD4040B with an AND gate on Q8, Q6 and Q3 to reset. That gives 99.902Hz but not 50% duty cycle. SN74LVC1G80 can divide Q8 by 2 yielding a 50% duty cycle clock at 49.951Hz. 

Hi Ron,

Thanks for the demo suggestion.

I'll check with my customer and see if they could use this. 

Thanks again.

James