How to shield a live wire at 240V ?

marconi · Apr 8, 2021

Introduction to RC Differentiator Circuits and Uses - https://www.bristolwatch.com/ele2/diff.htm

Passive differentiator circuit using a resistor and capacitor and waveforms.

marconi · Apr 8, 2021

q12x I had a go at putting together the circuits I posted this morning and then trying them out on version 1 of my IR proximity switch which is similar to your wings. I only made up two inputs to monitor the state of two of the yellow LEDs ( the 5th and 6th ones counting from the bottom). The differentiator detects the rising or falling edge of the switching pulse, and as wired on the breadboard at the moment it will detect and produce a beep only when an LED turns off. If you watch and listen carefully you can see how it performs.
The outputs of the comparators in version 1 fade and oscillate when the hand is moved slowly or remains still near the threshold points for LEDs 5 and 6. This affects the pulse waveforms which I highlighted in green and therefore the one which controls the 555 using the trigger input at pin 4 - you can hear the effect on the beep and illumination of LEDs 5 and 6.

In version 2 there is no fading or oscillation because I introduced some positive feedback/regenerative switching (hysteresis) at each of the 8 comparators, used some Schmitt trigger circuitry and employed a sample and hold technique which I described earlier ( S, P and Q). As an aside, the ruler LEDs are driven by the state of Q.

Much more work is required with the ideas I sketched out this morning to produce a click effect as good as in version 2. Alas, it will not be by me because I have achieved a satisfactory click effect in a different way and I don‘t have the motivation.

marconi · Apr 9, 2021

Vishay, BPV10NF IR Si Photodiode, 20 °, Through Hole 5mm package | RS Components - https://uk.rs-online.com/web/p/photodiodes/6997607/

q12x Good morning. I was looking for technical data on some 940nm ir leds and photodiodes I bought off Amazon only to discover that the photodiodes are in fact photo transistors according to several reviewers of the products so I will not be using them for my next sensor array.

I managed to find some 940nm photodiodes and 890nm LEDs on RS - see link and data sheet. This black photodiode has an integral daylight filter. Might be useful for your sensor too. There is a nice graph ( Figure 6 ) in its data sheet showing the relative sensitivity of the photodiode which shows it falling off rapidly towards the visible part of the spectrum.

marconi · Apr 9, 2021

I have spent a little while studying the way you actually build your modules and the components you use. I notice that you sometimes use glass body diodes for example in your SCM and in op amp circuit (3). Since your circuitry is exposed these diodes may be picking up and producing photo currents due to daylight or lamp light which interfere with or disturb the operation of the circuit. You might want to cover up these glass diodes and see if it makes any difference.

marconi · Apr 10, 2021

I have been pondering why LED1 is lit whenever one of the LEDs 2-10 is lit in dot mode. The LM3914 should not do this as you have noticed. For correct operation , the Vref voltage has to be less than or equal to the V+ voltage minus 1.5V.

You have two resistor connected in series between Vref and ground and their midpoint connects to the Vref adjust pin. If I remember correctly, the two resistors are the same value which means that Vref becomes 3Volts. This the voltage impressed across the internal voltage divider which provides voltage references to the 10 internal comparators which switch the LEDs on and off.

My first thought is that for correct operation V+ (pin 3) with respect to the V- (pin2) must be equal to or greater than whatever the Vref (pin7 ) to V-(pin 2) voltage is plus 1.5V. Is it?

If Vref V1 is 3V, then V+ to pin2 must be 4.5V or more. Is V+ decreasing below 4.5V when the LEDs turn on?

Take voltage measurement at the pins of the ic.

marconi · Apr 11, 2021

The way you have wired pins 2 and 4 of the LM3914 may also be causing a changing voltage drop between these pins and also with reference to the local 0V pad. I hope the attached diagram explains well enough that the circuit requires a local start point (X).

see:

https://www.ti.com/lit/ds/symlink/lm3914.pdf?ts=1614700327590&ref_url=https%3A%2F%2Fwww.ti.com%2Fproduct%2FLM3914

Page 18 Application Hints: Three of the most commonly needed precautions for using the LM3914 are shown in the first typical application drawing showing a 0V–5V bar graph meter. The most difficult problem occurs when large LED currents are being drawn, especially in bar graph mode. These currents flowing out of the ground pin (V- pin 2)cause voltage drops in external wiring, and thus errors and oscillations. Bringing the return wires from signal sources, reference ground and bottom of the resistor string (as illustrated) to a single point very near pin 2 is the best solution.

marconi · Apr 12, 2021

q12x Good morning. I thought of this idea to generate a click pulse whenever an LED turned on. (You would need a second op amp comparator to provide a pulse whenever an LED turned off, wired like the one I have shown but with its inputs (+ and -) reversed). The LM3914 would have to be in bar mode. The idea relies on the LM3914 LED outputs being constant current drive signals to each LED. Each time an LED turns on there is then a step up in current I through the common resistor which connects them all to Vled. Thus there will be produced across this resistor a step in voltage which the CR differentiator across it will turn into a short downward pulse V. During the downward pulse the differential voltage into the comparator will be less than 0V so the output will be 0V as shown by the waveform X. The X waveform can gate the click on and off.

marconi · Apr 12, 2021

q12x I wondered if you were still looking at this thread? I have just treated myself to an 'entry-level' ie cheap, simple to use yet useful - 2 channel digital oscilloscope as an early birthday present to myself (60 this year in June). I need it to pursue version 3 of my version of your project which tackles daylight and ambient IR.

This is the one I bought yesterday and arrived today:

_q12x_ · Apr 12, 2021

Yah, Hanmatek DOS1102 Oscilloscope is a very nice scope, and very nice presented. Thank you so much for it, mister @marconi . I will consider it for the future
This is a little toy I made. Well, it is an addon, an upgrade. It saves space!!! and it fill it's purpose VERY nicely, into my project as a Testing component.

Well, 2 toys. Haha.
- The new updates are awesome and great. With a full support and full help from my good american friend, mister Steve, we manage to make a very practical and very well constructed, very robust in other words, IR remote control and it's receiver. We actually tested it on the wings and is just perfect. I must admit, I like more the remote receiver output to S.5, its very linear and nice to watch. It's a very nice feeling. I will do a video soon about it.
I will start today to build the modules for this remote. Ive also made a range test from outside my room, so it's around 6meters until my testing table and it is working excellent. I can not be more happy about it. I am VERY happy about it.
- Another update is that I receive 10pcs of PIC12F508. I started already to swim in the jungle of tutorials and how to program them. I already have experience with programming PIC's from 1997 to 2003 or something. It's a long story. So its not my first time now, but it passed a lot of time from my tests back then. So I have to remember and remake those steps again. I have an awesome Pickit2 that is still working from those times and the tests so far are optimistic. We'll see. I dont want to talk too much about this subject since it can grow very ugly very quickly in 1000 branches of subjects that is not fun to keep track. Suffice is to say, that I started with it and I have to dig into it more until I can figure it to function optimal.
- The sound problem, the ticks. Mister @marconi , you did very good job and I referenced your work when I thought about this subject. Your doing, for me is an idea and a way to do it. So you helped my final decision even if is not the same as yours. I think I am more happy to build the sound ticks over another lm3914 in paralel with the existing one for the wing leds. I will basically have to drag 3 wires for its module: the signal to pin 5, from the other IC pin 5 as well, and the power pos and neg wires. Thats it. Very compact and robust.

The NEXT thing to think about it is the movement of the floating parts I mentioned some time ago. My thought is to use those micro-motors from vibrators in phones, I have 10 of them already and I hacked one to transform it back to motor from vibrator. Another idea is to use "smart wires", basically they are wires that change their shape when current passes through them, and come back to original state when not powered. The catch is they are very expensive. These are ideas for the moment. We will have to think more about this subject. And it is the final thing to add to the project.
These 3 remained - the remote, the bip sound and the movement. I am amazed of how it is turning out. And I am very confident it will look and work great in the end.

marconi · Apr 13, 2021

q12x I too am very pleased to read about your many accomplishments with this challenging project. I will look at the video as soon as you post it. I think you and your work demonstrate what Voltaire said:

No problem can withstand the assault of sustained thinking - philosiblog - https://philosiblog.com/2011/12/02/no-problem-can-withstand-the-assault-of-sustained-thinking/

I think the idea of using a second LM3914 is true genius - so simple, so reliable, so effective, so economical. (I wish I had thought of it!)

I cannot remember the floating parts feature. It will be interesting to follow the evolution of this design so I hope you continue to post. I am not at all familiar with the PIC12F508. I will look it up today.

I have made a an exact replica of my 850nm IR sensor array but using 940nm photodioes and 890nm LEDs. I have connected it to my version 2 and it does indeed appear less troubled by daylight and artificial lighting but not completely. I have not pulsed the 890/940nm sensor yet which is the major theme for Version 3. I thought I would study the signal from this sensor first using my oscilloscope for continuous and then pulsed IR. I want to 'see' the nature and magnitude of the interfering IR signals and the effect pulsing has.

I have not made up my mind yet about whether to have 8 comparators again examining the sensor signal acting as a flash (ie fast and the full voltage range at the same time continuously) analogue to digital voltmeter/converter or whether to only have one comparator and feed it with a ramped/stepped voltage reference and to then sample the comparator output at time/voltage intervals. The former would use something I know works; the latter would occupy less space but would require time and effort to design and test and be a distraction from the real challenge of tackling the interfering ambient IR problem.

Finally for now, I thought I would bring to your attention I discovered the sister ic to the LM3914 while reading up on the operation of the LM3914. The LM3914 is a linear a to d chip. There is a LM3915 which is logarithmic, that is to say an LED lights for every doubling (3dB change) in input voltage to pin 5. A log chip rather than a linear chip would help to even out the points when the LEDs light as the hand approaches - I achieved this by my 10 adjustable voltage references from the blue multi-turn trimmer pots and 10 comparators. The 3915 does the same internally.

https://www.electroschematics.com/lm3915-datasheet/

PS: Look up Bob Pease and Bob Widlar - I think you may be related to them!

marconi · Apr 13, 2021

The pickit2 reminds me of my introduction to microprocessors during my degree when we used a Motorola 8085 kit similar to this:

https://www.indiamart.com/proddetai...raining-kits-with-led-display-4341655888.html

We had to start off doing really basic things like input a word of 8 bit data, doing some mathematical or logical operation on it and then output it to turn on the appropriate Lights representing the output 8 bit word and then after so many seconds flashing on and off these lights.

About a hundred short exercises which gradually became more complicated. Happy days. I ended up using the 8085 to control a chip which compared in a special way ( cross correlation) two 16 bit words. It did not work until the nth version! I will keep n secret.

I commend this bite-size approach to exploring the jungle.

_q12x_ · Apr 13, 2021

This video is made for the IR remote control.
I talk here about the 3 remaining things I want to add, and also the motor(s) part.
I actually show them.

marconi · Apr 13, 2021

I have watched your video. Thank you for the update and remarks. Towards the end you talk about finishing off the design perhaps with some enclosure over the modules. Are you aware of ‘light pipes’?

See:

Buy LED Light Pipes | Essentra Components UK - https://www.essentracomponents.com/en-gb/p/led-light-pipes

There are other suppliers which may sell smaller numbers.

I asked my music minded wife the word to describe the beep sound you want to produce. It is an arpeggio.

_q12x_ · Apr 13, 2021

Thank you for watching it.
Why do you mention the light pipes? I am aware of their existence, but i did not know the name of them if i wanted to search them. So the name is new for me.
What should I do with them?

marconi · Apr 14, 2021

I thought you could put a cover over the board with the modules on. But that would obscure the LEDs. What you could do is use the light pipes to take the light from the LEDs from the wings moduleto holes in the cover so they could be seen. The light pipes have flat or rounded ends. You as an artist could then design and draw the wings on the cover around these light pipe exit holes.

_q12x_ · Apr 15, 2021

marconi said:
I thought you could put a cover over the board with the modules on. But that would obscure the LEDs. What you could do is use the light pipes to take the light from the LEDs from the wings moduleto holes in the cover so they could be seen. The light pipes have flat or rounded ends. You as an artist could then design and draw the wings on the cover around these light pipe exit holes.

One of the original objective for this projekt was to be as FLAT as imaginable. I originally wanted to be flat as a piece of paper or as a paint thickness. Thats like 10micro. But in reality, we have the leds height that add to the board thickness and the other components heights. As you can observe, absolutly no stand up component (or vertical placement), all are horisontal placed to diminish the thickness of this board as originally intended.
I think I understand your point. Because the other components height, I should add these tubes on the leds to get them more in front. Is this correct? Well, it is a nice idea, if I want to cover all the electronics and not use this transparent cover. But the thing is that I want to show all those wires there, all my work. I want to show it is a prototype, because it is.
Yesterday, ALL day, I built the receiver circuit Module. Today I finish it and it works ! Youhoo. Next, is to make the transmiter.

marconi · Apr 15, 2021

You have been more industrious than me. I have at least made a breadboard for version 3 which is bigger. You can see the 850/940nm ir sensor array.

_q12x_ · Apr 15, 2021

If you are happy with the functionality you get from the breadboard right now, it is time to make it on fiberglass board. It's probably a good advice for you, to make it as a definitive project. And dismantle everything you have here on these breadboards. I did the same. Remember the movie where I presented the remote control receiver? It was breadboarded. Now is gone, breadboard is clean again, ready for other experiments. But I made the receiver module in this time, unfortunatly with the same opamps on the breadboard because they were the only ones I had.

_q12x_ · Apr 15, 2021

mister @marconi - I am in need of your advice.
I have 2x555 to build my transmitter. They are SMD type. With very tiny pins on them.
Usually I cut like 3x3mm metal pad as the tiniest. Sometimes I go to 1x3mm but rarely. The thing is on such tiny surfaces, the solder is not behaving normally as on a pad. It is like a ball all around the tiny diameter. Also it is the problem that such tiny wires (not pads) are not very stable in my cardboard. I need some sort of glue on them as well. Maybe.
So...I need your advice, from what to build such tiny pads for a SMD IC pins? and how to stick them more firmly?
Any idea may lead to other ideas. Who knows. Until now i made like 2 or 3 smd IC circuits and it was hard as f**k. I can do it like that again, but... im out for ideas, maybe you may know an easier solution.
Thank you.