So recently I got some TS555 timers and some 1nF -100nF polyester capacitors and wanted to test them by designing an astable config that should oscillate at 200kHz (circuit diagram attached), I then built it on a breadboard. However I noticed that the frequency was too low, and figured it was the parasitic capacitances of the breadboard, so I ended up soldering it onto a PCB. It still however oscillated around 155kHz. So I measured the timing capacitance with my multimeter (quite cheap so very questionable accuracy) and it was around 0.977nF, which is not 1nF but it shouldn't produce a 50kHz error. I measured the resistances and they also were only slightly off. I measured the output pin floating as shown in the diagram and this circuit doesn't actually need to drive anything it was only for testing. Googling and browsing forums didn't give me any answers. I feel like maybe increasing the resistances by an order of ten and decreasing the timing capacitor by an equivalent amount to keep the same timing might solve the problem? Does anybody know what I am doing wrong? P.s find the datasheet for the timer attached.

 

Welcome to AAC!

Do you have an oscilloscope?

 

test them by designing an astable config that should oscillate at 200kHz

Interesting. My calculator says that circuit should oscillate at 200khz with a 58% duty cycle.
My test circuit reads 180Khz, which is within 10%.
Most likely the capacitors.

 

Interesting. My calculator says that circuit should oscillate at 200khz with a 58% duty cycle.
My test circuit reads 180Khz, which is within 10%.
Most likely the capacitors.

There is a problem here. If the capacitor is smaller than 1nF, that would make the charging and discharging times smaller, which in turn implies that the astable frequency should be higher. Your schematic does not indicate what supply voltage you are using. What is that value by the way?

 

Welcome to AAC!

Do you have an oscilloscope?

Yeah I got a Hantek DSO4102C which is what I used to measure the output signal.

Interesting. My calculator says that circuit should oscillate at 200khz with a 58% duty cycle.
My test circuit reads 180Khz, which is within 10%.
Most likely the capacitors.

I simulated my circuit in LTSpice with an NE555 model and it produced a signal around 180kHz to 190kHz aswell with a duty cycle around 54 iirc. So I was expecting some error. However 155kHz must indicate some error not related to tolerances right?

Your schematic does not indicate what supply voltage you are using. What is that value by the way?

5V

 

Yeah I got a Hantek DSO4102C which is what I used to measure the output signal.

Is the oscilloscope timebase timing accurate?

 

Is the oscilloscope timebase timing accurate?

Forgive me but I will have to look into that. I did however also measure the frequency with my multimeter to check if the oscilloscope was wrong but it showed 155kHz aswell. Also the oscilloscope correctly measures the 1kHz 5v square wave it produces for tuning the probes

 

So I rebuilt the circuit on my breadboard and found that the frequency was again around 155kHz so I can assume that the breadboard doesn't effect it much. I then replaced the two resistors with variable resistors so I could adjust them until I get the correct frequency and duty cycle. I managed to get 200kHz and 58% with the 1.2K resistor being 1.250K and the 3k resistor being 1.84k. I have built lower frequency astable circuits with the TS555 where the calculations were correct. So should I assume the equations for frequency and duty cycle of a 555 timer don't hold at higher frequencies and that I must use trial and error to get the correct timing values?

 

I'm out of ideas and still baffled by the evidence presented. Ordinarily I'd suggest another pair of eyes, but that too would have it's problems.

 

There is a problem here.

You were replying to the wrong person, I'm not the TS

 

So should I assume the equations for frequency and duty cycle of a 555 timer don't hold at higher frequencies and that I must use trial and error to get the correct timing values?

The equations are correct. They're based on the time it takes a capacitor to charge and discharge, so most of the variance you'll see is due to component tolerance and other characteristics.

Since you have a scope, use it to view the timing cap waveform. After the initial cap charge, it should fluctuate between 1/3 and 2/3 of Vcc.

 

Nobody notices that the TS555 is a CMOS 555, not a much more powerful NE555 or LM555. Then its output current is not high enough to drive the low value resistors, especially when its supply voltage is low.

 

You were replying to the wrong person, I'm not the TS

I never thought you were the TS, it is just a back and forth among several interested parties.

 

The datasheet for the Cmos LMC555 shows a circuit with a 200pF timing capacitor producing 3MHz but does not say the supply voltage.
The datasheet for the Cmos TLC555 shows a similar circuit with a 200pF timing capacitor producing a minimum of 1.2MHz (2.1MHz typical) but its supply is only 5V.
The datasheet for the Cmos ICM7555 shows the same circuit with a 200pF capacitor and a 5V supply producing 1MHz typically.

 

Nobody notices that the TS555 is a CMOS 555, not a much more powerful NE555 or LM555. Then its output current is not high enough to drive the low value resistors, especially when its supply voltage is low.

I do remember some "recommendations" on ranges for component values. The choices are not unlimited, and there is no unique solution of values for a given frequency.

 

The equations are correct. They're based on the time it takes a capacitor to charge and discharge, so most of the variance you'll see is due to component tolerance and other characteristics.

Since you have a scope, use it to view the timing cap waveform. After the initial cap charge, it should fluctuate between 1/3 and 2/3 of Vcc.

View attachment 233606

here's the output from the oscilloscope, the capacitor charging from 1.28V to 3.52V

 

The lower value is less than 1/3 of 5 (1.667) and the upper value is greater than 2/3 of 5 (3.333) so something is definitely amiss with the threshold values.

 

The lower value is less than 1/3 of 5 (1.667) and the upper value is greater than 2/3 of 5 (3.333) so something is definitely amiss with the threshold values.

Just measured control pin voltage, was 3.28v, seems fine. Why is it charging an extra 0.24V before going low hmm

 

Heres the cap voltage, with 3.28 and 1.64 labelled

 

Just measured control pin voltage, was 3.28v, seems fine. Why is it charging an extra 0.24V before going low hmm

Do you have a decoupling cap on the power supply? Have you measured it for ripple?