Can someone with experience using the NE556 help answer a question

[dnessett]

I am trying to create a one-shot using an NE556 timer IC. The circuit I have devised is shown in the first attachment to this message. There is nothing special about it. I just followed the directions in the data sheet. At the input is an RC differentiator (R2-C3) that I found out is required, since the NE556 one-shot will not terminate while the input is left low. The timing RC pair (R1-C1) is a 100K resistor and a .47 uF capacitor. According to the literature I read, the time constant should equal 47 mSec, which is how long the one-shot should last each time it is triggered. However, when I put a scope on the signal (see second attachment), the one-shot lasts only 1.5 mSec. Note that the oscilloscope trace shows the input after differentiation (the red trace) and the output (yellow trace).

I have computed the time constant several times and each time I get 47 mSec. I feared bring up this question, because I was worried that I am making some really stupied error. However, my frustration finally triumphed and hence this post. If someone with experience with the NE556 could help, that would be greatly appreciated.

[xrunner]

Are you sure R1 and C1 are the values you think they are? Did you try another set? Maybe you have a bad connection on a breadboard?

[Andy Watson]

At the input is an RC differentiator (R2-C3) that I found out is required, since the NE556 one-shot will not terminate while the input is left low.

But your circuit shows that the input (to the 556) is pulled low (to zero volts) by R2.
Where is the zero-volt reference on the your scope trace? To me, it looks like the input to the chip is being driven negative by a volt or more, it then settles-out at about -0.6v (one silicon junction voltage drop below zero volts). I would suggest that the 1.5mS is the time it takes for the chip to recover from having its inputs forced below the zero (ground) reference.

The minimum you need to do is change R2 to pull-up to Vcc such that the chip is not permanently triggered. You might then consider some means of preventing the input from exceeding Vcc (which might also have a negative impact on the chip).

[dnessett]

I have tried other combinations and they did not work as well.

[dnessett]

At the input is an RC differentiator (R2-C3) that I found out is required, since the NE556 one-shot will not terminate while the input is left low.

But your circuit shows that the input (to the 556) is pulled low (to zero volts) by R2.
Where is the zero-volt reference on the your scope trace? To me, it looks like the input to the chip is being driven negative by a volt or more, it then settles-out at about -0.6v (one silicon junction voltage drop below zero volts). I would suggest that the 1.5mS is the time it takes for the chip to recover from having its inputs forced below the zero (ground) reference.

The minimum you need to do is change R2 to pull-up to Vcc such that the chip is not permanently triggered. You might then consider some means of preventing the input from exceeding Vcc (which might also have a negative impact on the chip).

The differentiator has positive and negative spikes, the latter does go quite a bit below ground, since the capacitor turns the input into an AC signal with a ground base. I know with TTL there is an internal diode that protects the inputs from going more than .6V below ground. I don't know if that is also true for the NE556. I could put a diode after the differentiator clipping the input to stay below 5V (actually below 5.6V). See the attachment. Does that sound reasonable? If you have a better idea, please advise me.

[Andy Watson]

Yes, that looks better. Does it work?

This datasheet:
http://www.555-timer-circuits.com/datasheets/555an.pdf
shows that the trigger input is the base of a PNP transistor that has its collector in the ground. So it does effectively have a diode clamp to ground, however, this is not intended to be used as a clamp.

Also, what have you done with the other half of the 556? It may require its unsued inputs (especially reset) to be tied low, or inactive for reliable operation.

[dnessett]

Yes, that looks better. Does it work?

This datasheet:
http://www.555-timer-circuits.com/datasheets/555an.pdf
shows that the trigger input is the base of a PNP transistor that has its collector in the ground. So it does effectively have a diode clamp to ground, however, this is not intended to be used as a clamp.

Also, what have you done with the other half of the 556? It may require its unsued inputs (especially reset) to be tied low, or inactive for reliable operation.

It was a busy day today, so I will have to try the solution out tonight. Good point about tying the unused inputs to low. I will also do that. Thanks for the help.

[Zero999]

The RC time constant for the input circuit is quite long.

The positive spike can be eliminated by biasing the trigger at half the supply and connecting it to the input via a resistor equal to the values of the potential divider in parallel.

Here's an example circuit I designed awhile ago.


555 monostable edge 1ms.asc

[dnessett]

Yes, that looks better. Does it work?

This datasheet:
http://www.555-timer-circuits.com/datasheets/555an.pdf
shows that the trigger input is the base of a PNP transistor that has its collector in the ground. So it does effectively have a diode clamp to ground, however, this is not intended to be used as a clamp.

Also, what have you done with the other half of the 556? It may require its unsued inputs (especially reset) to be tied low, or inactive for reliable operation.

It didn't work. It appears the NE556 triggers on negatively going spikes (see attachment 1). The positive spike from the differentiator seems clamped below 5V. However, the one-shot pulse is reduced to ~440 uSec (see attachment 2).

[dnessett]

The RC time constant for the input circuit is quite long.

The positive spike can be eliminated by biasing the trigger at half the supply and connecting it to the input via a resistor equal to the values of the potential divider in parallel.

Here's an example circuit I designed awhile ago.


555 monostable edge 1ms.asc

Thanks. I'll give your circuit a try.

[floobydust]

RESET and TRIG are active-low, the pin labels should have a bar over them, dot on the pin etc. This makes it easier to understand the IC.
That said, I believe OP's pull-down R2 on /TRIG is keeping it constantly triggered - so it is timing the 'untriggered' interval based on the differentiator. C3 is too big I thing. Move R2 to +5V.

[dnessett]

The RC time constant for the input circuit is quite long.

The positive spike can be eliminated by biasing the trigger at half the supply and connecting it to the input via a resistor equal to the values of the potential divider in parallel.

Here's an example circuit I designed awhile ago.
<Image not shown. See previous post>

I used your trigger configuration (with my RC one-shot timing constant) and it worked. Attachment 1 shows the pulse generated by the differentiator with voltage divider per your design. Attachment 2 shows the ~47 mSec one-shot pulse. The differentiator pulse still goes below ground by a volt, but that does not seem to interfere with the one-shot generation. Thanks for the help.

[Zero999]

The RC time constant for the input circuit is quite long.

The positive spike can be eliminated by biasing the trigger at half the supply and connecting it to the input via a resistor equal to the values of the potential divider in parallel.

Here's an example circuit I designed awhile ago.
<Image not shown. See previous post>

I used your trigger configuration (with my RC one-shot timing constant) and it worked. Attachment 1 shows the pulse generated by the differentiator with voltage divider per your design. Attachment 2 shows the ~47 mSec one-shot pulse. The differentiator pulse still goes below ground by a volt, but that does not seem to interfere with the one-shot generation. Thanks for the help.

Great. I'm glad you got it working.

Do you have a USB stick/cable? If so, in future you can take a screenshot, to save you using a camera and the file size is also much smaller.

[dnessett]

The RC time constant for the input circuit is quite long.

The positive spike can be eliminated by biasing the trigger at half the supply and connecting it to the input via a resistor equal to the values of the potential divider in parallel.

Here's an example circuit I designed awhile ago.
<Image not shown. See previous post>

I used your trigger configuration (with my RC one-shot timing constant) and it worked. Attachment 1 shows the pulse generated by the differentiator with voltage divider per your design. Attachment 2 shows the ~47 mSec one-shot pulse. The differentiator pulse still goes below ground by a volt, but that does not seem to interfere with the one-shot generation. Thanks for the help.

Great. I'm glad you got it working.

Do you have a USB stick/cable? If so, in future you can take a screenshot, to save you using a camera and the file size is also much smaller.

Actually, I tried to do that, but couldn't get it to work. It didn't seem to recognize the USB sticks I used. Perhaps they were not formatted with the right file system.

[Zero999]

The RC time constant for the input circuit is quite long.

The positive spike can be eliminated by biasing the trigger at half the supply and connecting it to the input via a resistor equal to the values of the potential divider in parallel.

Here's an example circuit I designed awhile ago.
<Image not shown. See previous post>

I used your trigger configuration (with my RC one-shot timing constant) and it worked. Attachment 1 shows the pulse generated by the differentiator with voltage divider per your design. Attachment 2 shows the ~47 mSec one-shot pulse. The differentiator pulse still goes below ground by a volt, but that does not seem to interfere with the one-shot generation. Thanks for the help.

Great. I'm glad you got it working.

Do you have a USB stick/cable? If so, in future you can take a screenshot, to save you using a camera and the file size is also much smaller.

Actually, I tried to do that, but couldn't get it to work. It didn't seem to recognize the USB sticks I used. Perhaps they were not formatted with the right file system.

It needs to be FAT32 and under a certain size.

You can also connect it to your PC, but you need some software and it's not something I've done before.

[dnessett]

You can also connect it to your PC, but you need some software and it's not something I've done before.

The USB sticks I used were FAT32, so it must be the size. I looked at the OWON manual under technical specifications and could not find a size limit. However, all of the USB sticks I have are 16GB, which is probably too large for the scope.

[Zero999]

You can also connect it to your PC, but you need some software and it's not something I've done before.

The USB sticks I used were FAT32, so it must be the size. I looked at the OWON manual under technical specifications and could not find a size limit. However, all of the USB sticks I have are 16GB, which is probably too large for the scope.

I had forgotten, as it was while ago since I dealt with it, so had to do a search.

The problem is the cluster size is limited to 4K. Refer to page 51 of the manual.
https://fys.kuleuven.be/pradem/handleidingen/owon-smartds-series-user-manual.pdf

The supported format of the USB disk: FAT32 file system, cluster size cannot exceed 4K.

If you're running Windows, you'll probably need a third party formatting tool, otherwise use a smaller USB drive.

[dnessett]

You can also connect it to your PC, but you need some software and it's not something I've done before.

The USB sticks I used were FAT32, so it must be the size. I looked at the OWON manual under technical specifications and could not find a size limit. However, all of the USB sticks I have are 16GB, which is probably too large for the scope.

I had forgotten, as it was while ago since I dealt with it, so had to do a search.

The problem is the cluster size is limited to 4K. Refer to page 51 of the manual.
https://fys.kuleuven.be/pradem/handleidingen/owon-smartds-series-user-manual.pdf

The supported format of the USB disk: FAT32 file system, cluster size cannot exceed 4K.

If you're running Windows, you'll probably need a third party formatting tool, otherwise use a smaller USB drive.

Thanks.