When making analog op-amp or 555 timers, many times RC time constant is employed. Lets say we want 1 sec time constant: We could chose for instance the combination R=100k and C=10uF. Or we could choose R=100Meg and C=10nF. We have zillions of possibilities here.

If possible I was always choosing the combination such that I can avoid using electrolytic capacitor type. For example in my last circuit, to obtain a 10 sec RC time constant I used R=1Meg C=10uF. I would like to use nF level capacitors to avoid using electrolytic capacitor. But in this case then I have to increase R to very high resistances.

Here my question is:

1-) Is keeping the cap value low in nF level and increasing the R to hundreds of mega-ohms problematic? I thought we might be avoiding the use of electrolytic capacitor but then the very high resistor value could introduce more error?

2-) If I keep the resistor in kOhm level I should use then uF level cap. In this case what is the best alternative to electrolytic caps in uF level to obtain better accuracy and repeatability.

edit: Capacitors I use should not be surface mount.

  • 1
    \$\begingroup\$ When using a 555 timer, I'd always question why you're not using a mcu. Wins on total cost, precision, and gives at least 3 or 4 I/O minimum as far as I can tell. Valid question for op-amp though. \$\endgroup\$ – Sean Houlihane Oct 19 '16 at 10:51
  • \$\begingroup\$ its not my choice. i have to use analog way \$\endgroup\$ – user16307 Oct 19 '16 at 10:54
  • \$\begingroup\$ Oh, a homework question. \$\endgroup\$ – Sean Houlihane Oct 19 '16 at 10:55
  • \$\begingroup\$ No it is not is a task \$\endgroup\$ – user16307 Oct 19 '16 at 10:55
  • 1
    \$\begingroup\$ @user16307 Take a look at CD4060 or 74HC4060 chip. It includes a RC clock oscillator (as well as crystal) at the front end, and many stages of flip-flop asynchronous counters, used to extend the time delay. The RC clock would be short time-constant using small parts. The Reset pin can be used to make a one-shot timer. \$\endgroup\$ – glen_geek Oct 19 '16 at 14:54

Electrolytics have quite high leakage currents so they don't make a good choice and they also have a poor tolerance that gives significant variability in the design of a time delay circuit.

A commonly available non-polarized, 5% tolerance, surface mount type, I would consider is the X7R dielectric ceramic capacitor. In 1206 size you can get 10 uF 16V. If you are prepared to accept a 10% tolerance you can get a 47 uF in a 1210 case but it might only be 10 volts rated.

Worse dielectrics than X7R (such as X5R) yield higher values with wider tolerances and some change in capacitance with applied voltage thus making them unsuitable for decent timing applications.

So, depending on how accurate your timing needs are, I would make the assumption that 10 uF is the maximum (for half decent surface mount ceramic capacitors) and calculate the resistor based on this. If the resistance value is above 1 Mohm, I would start to worry about leakage currents from the pins on the chip and go another way entirely such as using a small MCU.

  • \$\begingroup\$ here is the circuit: i.stack.imgur.com/0hH55.png Im tring to set R1 and C4. But when I use 1Meg I face to use uF level caps which are electrolytic. I can live with 2 3 seconds error. Should I go for dielectric ceramic capacitor? \$\endgroup\$ – user16307 Oct 19 '16 at 12:34
  • 1
    \$\begingroup\$ The only worry is the 45nA bias current typically produced by the inputs of the LM324. Into 1 Mohm, 45 nA produces an error voltage of 45 mV which, as a percentage of the 9V rail is only 0.5% so I wouldn't worry about it. You could probably go to 10 Mohm and not notice much of a timing error. \$\endgroup\$ – Andy aka Oct 19 '16 at 12:54
  • \$\begingroup\$ sounds great. than for R4 resistor i might use one 5Meg resistor and one 5Meg poti to trim the delay. And a 1uF ceramic capacitor as C1. Or two 440nF and a 220nF cap in parallel if I cannot find 1uF ceramic. \$\endgroup\$ – user16307 Oct 19 '16 at 13:30
  • \$\begingroup\$ i found this one: uk.rs-online.com/web/p/ceramic-single-layer-capacitors/7215265 it says X7R dielectric. i think this is the one u suggest? \$\endgroup\$ – user16307 Oct 19 '16 at 14:04
  • \$\begingroup\$ I was only suggesting surface mounting components but that one looks OK. \$\endgroup\$ – Andy aka Oct 19 '16 at 14:12

If you can live with the limitations of the 555:

1-) Is keeping the cap value low in nF level and increasing the R to hundreds of mega-ohms problematic?

Yes, it is problematic. There is a minimum current required to trigger the threshold comparator in the 555. Typical 555s list the maximum value as 0.25 uA (250 nA) - check your specific 555 datasheet in case it is different. This sets an upper limit on the timing resistor value, which varies depending on the IC's power supply voltage.

Although devices will typically require less than 250 nA to trigger, using a lower current can result in unreliable operation.

For example, from the TI 555 datasheet:

Threshold current specification - from TI datasheet

Upper limit on timing resistor values - from TI datasheet

As explained in that excerpt from the TI datasheet, that requirement sets an upper limit on the resistor values - it says:

This parameter influences the maximum value of the timing resistors \$\small R_A\$ and \$\small R_B\$ in the circuit of Figure 12. For example, when \$\small V_{CC} = 5\;\textrm{V}\$, the maximum value is \$\small R = R_A + R_B ≉ 3.4\;\textrm{M}Ω\$, and for \$\small V_{CC} = 15\;\textrm{V}\$, the maximum value is \$\small 10\;\textrm{M}Ω\$.

Although they mention "Figure 12" in that footnote in the datasheet, which is the typical Astable configuration, the same limitation applies to "Figure 9" which is the typical Monostable configuration. I've included both here for ease of comparison, and marked the resistors which have that maximum total value referred to in the datasheet:

Monostable and Astable 555 configurations - from TI datasheet

Also note that some characteristics like accuracy, may be quoted within a specific range of timing resistor and capacitor values.

2-) If I keep the resistor in kOhm level I should use then uF level cap. In this case what is the best alternative to electrolytic caps in uF level to obtain better accuracy and repeatability.

Look for capacitors which have: Low leakage; low temperature coefficient (if important in your application); tight tolerance. These all affect "accuracy and repeatability".

  • \$\begingroup\$ im using lm324 : i.stack.imgur.com/0hH55.png trying to set R1 and C4 to 10 sec. But when I use 1Meg I face to use uF level caps which are electrolytic. btw I can live with 2 3 seconds error for 10 sec time constant. i couldnt find an alternative cap which is not surface mount. \$\endgroup\$ – user16307 Oct 19 '16 at 12:51
  • \$\begingroup\$ @user16307 - (a) "trying to set R1 and C4 to 10 sec" - I guess you meant R5 and C1 - there is no C4 in your circuit and R1 is not in a relevant location in the circuit. (b) You originally asked for "best alternative to electrolytic caps [...] to obtain better accuracy and repeatability", but now you "can live with 2 3 seconds error for 10 sec time constant", so accuracy is not required. "i couldnt find an alternative cap which is not surface mount." - Tantalum still avail in leaded package. Note: You are now changing your question e.g. must not be surface mount. Please update question :-) \$\endgroup\$ – SamGibson Oct 19 '16 at 13:18
  • \$\begingroup\$ oh sorry i would have written R4 and C1. yes not much accuracy but not too bad. i can live with 2 3 sec error but not 50% error when repeated after a year. that would be disaster. \$\endgroup\$ – user16307 Oct 19 '16 at 13:25

When resistors much above 10Mohm are called for, you need to start worrying about (or at least considering) board surface leakage currents, and the leakage current of electrolytics, if you weren't already worried about IC input bias currents.

For times much above 1 second, consider using an MCU (especially if you want to do other fancy stuff as well), or a 4040 or 4060 oscillator/divider IC, which is made for just such long time delay purposes.

  • \$\begingroup\$ precision is not too important serting 10 sec can result between 8 12 seconds. I just thought maybe elrctrolytic is worse than that since it has bad fame \$\endgroup\$ – user16307 Oct 19 '16 at 11:40
  • \$\begingroup\$ alli electrolytics do have a significantly worse leakage current than other types, and also lousy tolerances, best avoided in timing circuits if you want better than 50% repeatability/accuracy. \$\endgroup\$ – Neil_UK Oct 19 '16 at 13:13

At the 10uF range you can use a (relatively bulky and expensive) film capacitor and get reasonable tolerance. This is the old-school way of doing it. Mylar (polyester) is acceptable, some other films have advantages in some situations because of lower dielectric absorption. For cheapness, a special low leakage +/-20% tolerance electrolytic is another option.

You can try an X7R or X5R 10uF cap as @Andy suggests but keep in mind the capacitance may drop significantly as the voltage increases, and that drop may vary from manufacturer to manufacturer and even part-to-part. To minimize that effect, get a physically large part and derate the voltage significantly from the maximum.

Edit: Some caps such as some 50V X5R caps appear like they might be suitable, but you really have to pay attention to the details for the exact part number- here is one characteristic curve (typical)- this one is a 1210 so not even all that small: enter image description here

This particular X7R part basically disappears when you have more than about 2/3 rated voltage across it, and it's none too impressive even at 20% of rated voltage.

A microcontroller using only its internal RC oscillator can give you 1%-ish precision with virtually no external parts. It's certainly the solution of choice in most cases in the current era.

  • \$\begingroup\$ im not allowed to use micro-controllers. how about using ceramic caps in parallel instead of using a single electrolytic cap? i will consider mylar if it is not too expensive \$\endgroup\$ – user16307 Oct 19 '16 at 13:34
  • \$\begingroup\$ Putting more parts in parallel won't necessarily help the voltage coefficient. You have to read the datasheets. Mylar is falling out of favor a bit, but in quantity the dipped ones are fairly cheap and the box type ones a bit more expensive. \$\endgroup\$ – Spehro Pefhany Oct 19 '16 at 19:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.