3

This is a circuit I made to create a high signal when a switch is opened and closed. The duration of the signal can be adjusted by changing the capacitors and resistors.
It is also low powered when using a high resistance on R1 and R2.

This is a drawing and simulation I made on falstad.com/circuit enter image description here

Is there a name for this kind of circuit?

Note:
I edited this question since it was of poor quality, so this question is only about trying to name the circuit.
I have created a new question with all the details, experience and requirements that I have. Improving edge detector with latching circuit for esp8266

John-Arvid
  • 135
  • 9
  • Does this answer your question? [Falling and Rasing Edge Detector](https://electronics.stackexchange.com/questions/270089/falling-and-rasing-edge-detector). Ditto [this one](https://electronics.stackexchange.com/questions/270894/dual-edge-detector/270903#270903). And [this one](https://electronics.stackexchange.com/questions/374799/control-of-transistor-as-switch-at-power-up-and-power-down/374802#374802). – Andy aka Aug 15 '20 at 10:32
  • 1
    Thanks for your input, they are giving some input, but not something that I think is usable. For example, not handling ESD with pulldown resistors which I have used here. Also the hex inverter is not usable for me because of the voltage and current consumption, it also creates a more complex circuit without much benefit. (But these are my thoughts that are from a beginners perspective) – John-Arvid Aug 15 '20 at 15:48
  • Thanks for adding some detail. But there's no description of the application: why the low power, what's the switch from and why, what's the pulse for and why, how many are you making and so on. Without that, we can't guide you well in your stringent current requirements. – TonyM Aug 18 '20 at 15:54
  • Thank you for your follow up. I did not want to add that kind of information since I wanted it to be a more general question. I'm not trying to be difficult, just want to discuss. Low power: It is battery operated, and I think in general, the lower power usage the better. Switch: Reed, or any other, should not matter, could give a high or low signal. Pulse purpose: To wake up a MCU that is off 99% of the time. How many: Why does it matter? I want to make something easy for the DIY community. I can update it with requirements, but I wanted this to be a general thing. – John-Arvid Aug 18 '20 at 18:36
  • Sorry, John, your comment comes across as secretive and it's a Q&A site, not an I'll-reveal-more-as-we-go-if-I-have-to discussion forum. But there's too little for me to offer more technical help than I have unless you detail a good question. e.g. how many - 'what does it matter?'. Knocking one up for your garden shed door lock is a lot different to designing a 5k/month application, in costs and required reliability. I notice you asked almost the same question last year and that seemed to tail off, too. Humour me: add all the requested info in detail :-) In the meantime, I'll leave it there. – TonyM Aug 18 '20 at 19:10
  • No I'm sorry, I start to see that my train of thought has failed a bit here. Not trying to be secretive, I have the experience from programming that a good question is a more general one which the solution could be used in almost any similar situations (maybe not with electronics). Yes my previous question was of similar matters, but I wanted to update that one with my new knowledge from my experience and knowledge from this question. But it looks like I have failed asking a good question. So I will change this question later to a specific one with a different topic and tags. – John-Arvid Aug 18 '20 at 19:55

1 Answers1

5

It's an edge detector.

'out' pulses logic high when the switch is opened or closed. The pulse period is determined by C1, R2 and R3.

R1 pulls the upper XOR gate input logic low when the switch is open. D1 blocks the C1 voltage from reaching R1 at that time.

For a practical circuit, the resistor values are very high. The input leakage values of the logic gates are usually microamps so a 10M pull-down may well not reach the logic low voltage. That leakage current will affect the timing constants derived from R2//R3 and C1 when on a switch press and of R3 and C1 on a switch release.

Similarly, the reverse leakage current of D1 may become be enough to produce a significant voltage drop across R1.

As a rough and unsubstantiated guide, I would expect these resistor values to be all less than 100K in a real design. That may make C1 impractically large but that depends on the (undefined) application requirements. However, you can use a small C1 value to generate a short 'out' pulse that triggers a better monostable producing a longer final pulse.

TonyM
  • 21,742
  • 4
  • 39
  • 62
  • Thanks for your explanations. But if I would use a resistor value less than 100K for R1 and R3 the power consumption would be over 80 microamps. How could I avoid that? Just a disclaimer, I have built a working prototype with this setup witch I don't have any issues with yet. – John-Arvid Aug 15 '20 at 15:35
  • @JohnArvid, the target is not designing a circuit that works, it's designing a circuit that never doesn't work :-) The effect of input leakage current will show in variances across a large number of units built and with changes in temperature, supply rail and part aging. How much do variations matter in your application? Does your circuit timing accurately match the calculated behaviour of it? Hard to measure within the circuit as putting a scope probe or DMM across it is adding a 10Mohm or suchlike parallel resistance to it that'll disrupt it. Can you add exact part numbers to your schematic. – TonyM Aug 15 '20 at 16:52
  • hehe, I can relate to that from when I am working with programming, but since I am a novice regarding electronics I can't think about all the possible issues that could/would appear. So thank you for that note :-) For my application the only requirements are < 10uA, signal length > 300ms and simple components so I wouldn't need a new special programmer for a MCU. But I wanted to keep this out of the question to be more general, but I understand that is difficult. I don't have any equipment to compare measure to calculated. I will try to add a list in 24 hours. – John-Arvid Aug 15 '20 at 17:33
  • I think I will only add the part list here for now. The resistors are cheap 1/4W troughole. C1: 1uF 50v (electrolyte) Diode: Diotec 1N4004 XOR: SN74AHC1G86DBVR – John-Arvid Aug 16 '20 at 14:33
  • @John-Arvid, please all this info into your question, don't leave it in several comments. Otherwise you're expecting readers to piece together the actual question from all over the place. Thanks in advance. – TonyM Aug 16 '20 at 18:22