I have a strange phenomenon using NE555 in astable mode.

Here are the tests:

With NE555 ST or Signetics : no problem

With NE555 Texas Instruments :
Appearance at the output 3 of a "spike" of about 0.5 V on the top of the rising edge.

Here are the tests/modifications made but without success :

- Tests from 50 to 13 kHz and 4,5 V to 10 V
- Power supply decoupling with 100 micro and 1 micro and 100 nF
- Tests with different capas technologies on pin 2 (LCC, ceramic, ...)
- Test with increasing the resistors/decreasing capa values for the same frequency
- Tests with about ten TI chips

Attached are the schematics and scope with the comparisons
Would you have an idea ?

 

Is your pin 3 floating in the breeze, or is it loaded by something not on your diagram?

The spike, if you zoom in a bit in time, is ringing, which is some inductance and some capacitance forming an LC tank (breadboard stray inductance or capacitance, a wire connecting the probe, etcetera). If you put a bit of load on the output, that may flatten that ringing out.

If you are bypassing the IC by putting capacitors on the rails and not near the part, I would put them near the part. From VDD to capacitors to GND pin is the shortest path, from the 555's perspective.

OH, also. I would hook your probe up to your calibrator output signal on the oscilloscope and ensure that the probe is adjusted properly for compensation for just these specific situations. That occurred to me after I typed the first part. If you are unfamiliar with what I mean, let me know, and I or someone can explain.

Maybe these ideas will help a bit.

Paul
KI5VNH

 

Perhaps the chip from TI is made with newer technology with faster transistors. This can cause the formation of a sharper edge. Try putting 200 pF on the output. Or the same capacitor and resistor in series.

 

Is your pin 3 floating in the breeze, or is it loaded by something not on your diagram?

The spike, if you zoom in a bit in time, is ringing, which is some inductance and some capacitance forming an LC tank (breadboard stray inductance or capacitance, a wire connecting the probe, etcetera). If you put a bit of load on the output, that may flatten that ringing out.

If you are bypassing the IC by putting capacitors on the rails and not near the part, I would put them near the part. From VDD to capacitors to GND pin is the shortest path, from the 555's perspective.

OH, also. I would hook your probe up to your calibrator output signal on the oscilloscope and ensure that the probe is adjusted properly for compensation for just these specific situations. That occurred to me after I typed the first part. If you are unfamiliar with what I mean, let me know, and I or someone can explain.

Maybe these ideas will help a bit.

Paul
KI5VNH

Hi Paul

Thank you for these nice suggestions

1- Yes pin 3 is floating for tests. @Bordodynov (See below) suggested to put a 220 pF on pin 3 ...
2- I am using a breadboard. As I am a Ham like you I do understand what you mean !
3- The purpose of the circuit is to design a servomotor tester with 2 NE555.
4- OK I'll test the probe as you suggested it

Let me try and I'll keep you posted
My best 73
Philippe
F1 CUJ

 

Perhaps the chip from TI is made with newer technology with faster transistors. This can cause the formation of a sharper edge. Try putting 200 pF on the output. Or the same capacitor and resistor in series.

Thank you for this suggestion. I'll try it immediately with a 220 pF.

Yes I was upset because older designed chips from ST and Signetics have a nice signal ...

 

Thank you for this suggestion. I'll try it immediately with a 220 pF.

Yes I was upset because older designed chips from ST and Signetics have a nice signal ...

If you put one of those older chips in and zoom into that edge in time, I bet you'll still have distortion there from the same parasitics that I mentioned in the previous post...and you're welcome!

Best regards,
Paul KI5VNH

 

Try your tests using CMOS 555-timer such as LMC555, TLC555, ICM7555.

 

If you put one of those older chips in and zoom into that edge in time, I bet you'll still have distortion there from the same parasitics that I mentioned in the previous post...and you're welcome!

Best regards,
Paul KI5VNH

Paul,

I have first tested the probe. I have 2 different one. They are OK.
Then :
Test 1 : I have loaded pin 3 with a 470 Ohms and a LED. The spike remains
Test 2 : I put a 220 pF on pin 3 : nealy no effect
Test 3 : I have replaced the TI chip with a ST and a Signetics. The signal is clean and perfect

Any other ideas ?

 

Thank you for this suggestion. I'll try it immediately with a 220 pF.

Yes I was upset because older designed chips from ST and Signetics have a nice signal ...

Try your tests using CMOS 555-timer such as LMC555, TLC555, ICM7555.

Hi
thank you for your suggestion.
I don't have these CMOS version and I'll order some

 

The rows or contacts and wires all over the place on a solderless breadboard Guarantees capacitance, inductance and signal coupling that messes up signals.

 

Test 3 : I have replaced the TI chip with a ST and a Signetics. The signal is clean and perfect

Looks like it could be ringing to me. Try putting 500-100 ohms in the ground lead of the scope to dampen the oscillation.

 

The scope seems to show the top of the spike at 5.2V and the flat level at 4.8V. What is the exact voltage of your supply?

 

Are you using a 10x attenuation oscilloscope probe?
If not, get one and learn how to use it.

 

Power supply decoupling was mentioned above. It is critically important. Every operational parameter on the datasheet assumes proper decoupling. Start with a 0.1 uF ceramic cap with the shortest possible leads across pins 1 and 4. To this, add an aluminum electrolytic cap in parallel, something in the 10 uF - 47 uF range.

Another suggestion, change the timing components by a factor of 10. 10K resistors and a 0.1 uF cap.

 

Paul,

I have first tested the probe. I have 2 different one. They are OK.
Then :
Test 1 : I have loaded pin 3 with a 470 Ohms and a LED. The spike remains
Test 2 : I put a 220 pF on pin 3 : nealy no effect
Test 3 : I have replaced the TI chip with a ST and a Signetics. The signal is clean and perfect

Any other ideas ?

I was going to request an image of the setup and probe hookup. If you’d like to post that for our information, it couldn’t hurt.

I maintain that the phenomenon we speak about is occurring, if only to a lesser extent, on all chips. If you zoom in in time at one of the other older chips, you’ll see it, just not as badly.

It is caused by parasitic oscillation, and in order to minimize it, you must make the Q of this LC parasitic very low so that it does not ring.

If it is breadboard/lead related, close and multiple bypass capacitors should help. Short leads to the properly compensated scope probe should also help.

I am not surprised the loading didn’t help. That’s adding more R than it is L or C, and most likely in the wrong place.

The chips, by the way, have slight differences in their output circuitry…along with, as suggested, more modern IC fabrication techniques (possibly, I personally doubt it).

Hope that helps.

Paul
KI5VNH

Desole pour any spelling errors, typed on my phone

 

I will share my experience. I used a push-pull repeater. The supply voltage was 12 V. To my surprise the output of the repeater had a voltage spike up to 16 V. As it turned out the reason was the long power wires. After the power of the repeater was shunted with a ceramic capacitor the spike disappeared. A 0.1 uF ceramic capacitor was close to the repeater.

 

I would suspect that what you see is a flyback pulse after the 555 shorts out the power supply as it changes state. The short stores energy in the inductance of the wiring and is released when the short is removed causing a flyback pulse.
I think you have managed to prove that your SGS-Thomson and Signetics parts don't have as bad a power-supply short on change of state than your Texas part does. Interesting to know, but it doesn't necessarily mean that all SGS-Thomson parts are better than all Texas parts in that respect.

 

I will share my experience. I used a push-pull repeater. The supply voltage was 12 V. To my surprise the output of the repeater had a voltage spike up to 16 V. As it turned out the reason was the long power wires. After the power of the repeater was shunted with a ceramic capacitor the spike disappeared. A 0.1 uF ceramic capacitor was close to the repeater.

So the reason for the ringing (spike) was stray inductance…sounds familiar…

Good to know you found its source!

Paul
KI5VNH

 

Hi

I have finished the servomotor controller with 2 NE555.
I start a new thread in 5 minutes.

Thank you @PaulEE @Bordodynov @MrChips

 

I would suspect that what you see is a flyback pulse after the 555 shorts out the power supply as it changes state. The short stores energy in the inductance of the wiring and is released when the short is removed causing a flyback pulse.
I think you have managed to prove that your SGS-Thomson and Signetics parts don't have as bad a power-supply short on change of state than your Texas part does. Interesting to know, but it doesn't necessarily mean that all SGS-Thomson parts are better than all Texas parts in that respect.

Hi @Ian0 Thank you for the explanations. Very instructive
I have just finished the design of the 2 NE555 servo motor controller and 2 other problems remain !
I am just posting a new thread