R/C 5 minutes timer, how accurate can it be made?

[neazoi]

Hi I am thinking of a timer that uses an R/C to drive the base of a BJT, so as to create a 5 minutes delay.

How accurate this RC can be made when placed in a clock circuit with 5 minutes step (i.e clock minutes needle changes every 5 minutes)

I am thinking of using an NP0 capacitor (so it has to be small) and a very large 0.1ppm resistor. How accurate can it be in a 24 hour period?

Or another example, according to this calculator for example https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-time-constant a 50nf and 20Mohms will give a 1pps. How accurate this can be if using 0.1ppm resistors and NP0 capacitors?

 

[KlausST]

Hi,

Since tau = R × C... both devices errors have same influence.
When you use a 1ppm resistor you should use a 1ppm capacitor, too.

Klaus

 

[neazoi]

Hi,

Since tau = R × C... both devices errors have same influence.
When you use a 1ppm resistor you should use a 1ppm capacitor, too.

Klaus

Why is that KlausST?
What is a typical ppm for a NP0 capacitor?

 

[KlausST]

Hi,

Why is that KlausST?

Tau = R x C should be well known.
Increase R by x% and you get Tau increased by x%, too.
Increase C by x% and you get Tau increased by x%, too.

What is a typical ppm for a NP0 capacitor?

I'd read datasheets or consult the selection tools of manufacturers.

Klaus

 

[FvM]

The question is rather unclear. Do you consider to trim the time constant so that temperature drift and aging come into play or are you asking for initial accuracy with off-the-shelf parts? If the timer is build with BJT transistors, base current and other circuit non-idealities will ruin the theoretical RC accuracy. The huge gap between 1s and 5 min suggests that you consider frequency dividers (counters), if so why no go for a crystal oscillator?

 

[dick_freebird]

Capacitors tend to have both lousy tolerances
and high tempcos. Resistors can readily be had
at 1% and low TC in metal film.

The base impedance of a BJT would alter the
explicit R value and this also has significant
"make" (low current beta) and temperature
variation.

Accuracy would be better using a more complex
circuit. Such as an XO with following counter. If
I recall there was even a CD4060 logic part
with this kind of scheme, like 14 bits worth of
counter? Then you would be down to the XO
temperature stability and not much else.

 

[Easy peasy]

clock and divider a much better option ... e.g. CD4051

 

[stenzer]

Hi,

by mentioning a time of 5 minutes a large electrolytic capacitor comes into my mind, as al large time constant is needed. As already mentioned by @dick_freebird those capacitors and especially lectrolytic ones have large tolerances. I also would recommend a logic approach, the HEF4541B [1] might be handy for your application, as it provides a couple of features. It still utilzes capacitors and resistors, but much lower values are rquired and a NP0 ceramic capacitor might be used as well.

If it comes to accuracy, the ambient temperature change might be considered as well, have a look on page 12 in [1]. Keep in mind, also the base-emitter voltage is affected by the ambient temperature, thus controlling the BJT by an analog signal like your RC waveform leads to an altered switching time.

[1] https://assets.nexperia.com/documents/data-sheet/HEF4541B_Q100.pdf

BR

 

[d123]

Hi,

Very inaccurate. And, how do you measure the 20M resistor with any certainty?

50nF is 'huge' for accurate timing. 100pF already introduces error for 1ms with reliable repeatability. A frequency counter for a 1 kHz RC clock should show anywhere up to 10% error at one ambient temperature. Imagine accumulated error over 24 hours...

You could read comparison charts of capacitor dielectrics regarding capacitance change with applied voltage and temperature fluctuations and roughly extrapolate notions of expected minimum error. Polystyrene, polycarbonate, PPS, silver mica are good timing choices, NPO is okay, new silicon-based caps also allegedly.

Air wiring would remove invisible enemies like stray capacitance and inductance.

In my opinion, 1 second is already unrealistic for an RC network used in a clock, especially with a BJT and its temperature dependence.

I'd make the tau as short as possible, a second is a long time to make timing repeatably accurate for any capacitor, and divide down with counters, etc.

 

[BradtheRad]

Perhaps your project can pick up mains hum (via power supply or sensor)? Then send it through a counter which divides down the frequency to give you 5 minute intervals. (The count is 15,000 = 50 x 60 x 5.) It can be done with a 4020 IC or 4060.

 

[Easy peasy]

Oops ... CD4541, clock and divider ...