0
\$\begingroup\$

I have a 555 timer connected to an LED that I am trying to strobe for 450ms on, 72ms off to replicate an existing strobe light I have. The issue I am having is it is strobing at around 243ms on, 42ms off. According to my calculations, in order to get that timing I would need Ra=37.2k, rb=6.2k, with C=9.3uF. My actual values are Ra=56K, Rb=10K and C=9.3uF, however. According to the formulas found online for Th and Tl, with those values, I should be getting 460ms on and 69ms off, which is close enough for me. I am reading the pulse durations with a light sensor connected to an arduino running on a separate power supply. I've tried replacing the 555 timer with a different one (I bought 5) and it doesn't change anything so I'm really not sure what is happening to make it run the way it is. Is there something wrong with my circuit or am I missing something here?

Edit: the capacitor is a 106 ceramic, which should be 10uF but I measure it at 9.3uF

enter image description here

\$\endgroup\$
8
  • \$\begingroup\$ 9.3uF? That's not exactly a standard value. \$\endgroup\$
    – TimWescott
    Commented Jan 14, 2021 at 21:54
  • \$\begingroup\$ The capacitor I'm using is supposed to be 10uF but it measures at 9.3uF. \$\endgroup\$
    – joshua0823
    Commented Jan 14, 2021 at 21:54
  • \$\begingroup\$ If you want accurate timing (better than 5% accuracy), don't use a 555 instead use a microcontroller. \$\endgroup\$
    – Bimpelrekkie
    Commented Jan 14, 2021 at 22:00
  • \$\begingroup\$ The accuracy I'm currently getting is 50% off. is a 555 timer really that inaccurate? \$\endgroup\$
    – joshua0823
    Commented Jan 14, 2021 at 22:04
  • \$\begingroup\$ @joshua0823 capacitors, especially large ones like 10 µF, are far from ideal: their capacitance changes with bias voltage. So, this is pretty much doomed from the beginning; you won't achieve the accuracy you want. Since a NE555 is power-hungry, inefficient, complex to tune: go for a microcontroller, this is more effort than it's worth. \$\endgroup\$
    – Marcus Müller
    Commented Jan 14, 2021 at 22:05

2 Answers 2

Reset to default
2
\$\begingroup\$

Firstly, the ceramic capacitor will likely be a LOT less than 10uF with bias on it, as the voltage across it increases it will drop in value. It would not be unusual to see a small capacitor lose 50% or 70% of its nominal value.

Secondly, you're drawing a great deal of current from the power supply (LED with no series resistor) and there is no bypass capacitor on the 555 pin 5 and negligible capacitance across the supply (only 10nF). NE555 (bipolar) even on their own draw big surges when they switch.

Try bypassing the supply with 10uF ceramic, bypass pin 5 with 10nF or 100nF, and add a second regulator and second set of batteries just for the LED.

Or use a microcontroller, even with the internal RC clock you'll get infinitely more predictable performance.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks. I'm just going to use a microcontroller to save myself a headache \$\endgroup\$
    – joshua0823
    Commented Jan 15, 2021 at 7:46
0
\$\begingroup\$

According to my calculations, in order to get that timing I would need Ra=37.2k, rb=6.2k, with C=9.3uF. My actual values are Ra=56K, Rb=10K and C=9.3uF, however. According to the formulas found online for Th and Tl, with those values, I should be getting 460ms on and 69ms off, which is close enough for me.

Well, there's your problem. You are doing your calculations wrong. Apparently you have two different ways of calculating: your original way and the online formulas. So why not present both, and show your workings. Comparing the two should be pretty simple.

Edit: the capacitor is a 106 ceramic, which should be 10uF but I measure it at 9.3uF

Sounds about right. Assuming it's a 10% tolerance unit, anything between 9 and 11 uF is acceptable.

\$\endgroup\$
2
  • \$\begingroup\$ I can't find a calculator online where you can plug in the Th, Tl, and C and it spits out what Ra and RB should be. I just rearranged the formulas Tl=0.693*Rb*C and Th=0.693*(Ra+Rb)*C' to solve for Ra` and Rb with the actual Th, Tl, and C values I have. Are these values not correct? \$\endgroup\$
    – joshua0823
    Commented Jan 14, 2021 at 23:28
  • \$\begingroup\$ @joshua0823 - While you haven't said so, I'm going to assume that you've built a circuit using 56k, 10k and 9.3 uF, and are getting about the pulse widths you need. What you seem to be asking is why the data sheet equations are giving a different answer. The first thing I'd notice is that your circuit would just about work with the data sheet equations if the cap were 6.2 uF rather than 9.3. This suggests that your cap is behaving oddly, or that you have a bad connection to it. You need to check this out. Also, you will find EE life much easier if you by a cheap oscilloscope. \$\endgroup\$
    – WhatRoughBeast
    Commented Jan 17, 2021 at 1:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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