Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

frequency divider 2:1 reed contact & electro magnet

Status
Not open for further replies.

kael

New Member
I like to alter the triggering of a reed contact, so that this contact only gets halve of the original pulses coming from a magnet on a turning wheel. Dependig on the turning speed, the pulses are about 1 to 6 Hz.
I was thinking of adding an extra reed contact for this wheelmagnet and some circuit (555 based?) that only 'let through' every other pulse, so I get a 2:1 ratio.
So after 2 pulses at the new reed contact, an output is triggerd to activate an electro magnet, which in his turn triggers the original reed contact.
This circuit from the new reed contact upto the electro magnet, is in fact placed 'in between' the original magnet and original reed contact as a matter of speaking.

So my questions:
1. Who knows a suitable simple diagram to realise this, preferable working on battery (9V)?
Any idea on battery live? Does is it need a on/off switch to save battery when not in use?
2. It only needs to be a small electro magnet for the reed contact, I didn't find any ready available online, maybe I can use the magnet from a buzzer? These exist in ranges 1.5-12VDC, 15-30mA and are quite small.
3. Or maybe some whole other approach is possible, without changing the origal reed contact input though?

I found a similar circuit http://www.ecircuitslab.com/2012/05/tachometer-pulse-divider.html
but can this one handle the input reedswitch and output electro magnet?
 
Let me get this straight.
A magnet on a wheel actuates a reed switch.
You want a 9V circuit which gives 1 output pulse for every 2 actuations of the reed switch? If so a CD4013 will probably fit the bill. Cheap and consumes almost zero power.

BTW, many buzzers don't have electromagnets in them; they're piezoelectric devices.
I'm curious as to why you want to use an electromagnet to re-actuate the reed switch :confused:
 
Last edited:
Well I think I will need some kind of circuit to do this, so it need to be (small) battery powered, 9V looks my best choice but 1,5V is also OK.
I wasn't aware of the Cmos CD4013, do you think is suitable for this?
Yes I know it should be a magnetical buzzer not a piezo, to be able to use the magnet.
I don't see another way to actuate the new reed switch, besides an electro magnet who get its pulse from the 2:1 circuit.
I made a drawing of my plan:
On the right wheel you can see the modification with an extra reed added still using the orginal magnet. The electro magnet is yellow and actuates the original reed now.
I hope you have better view on my toughts now.
View attachment 65333
 
Last edited:
Per Alec just use a CD4013 Dual D Flip Flop (you only need one of the dual flip flops on the chip) configured as a /2 (divide by two). For every two pulses in you get one pulse out. It will run just fine on 9 volt battery power.

Ron
 
OK the CD4013 is fine then, but can the CD4013 handle directly the reed switch as input? Or does it need some (bounce) cleaning?
What about the output, I suppose I also need a transistor capable of powering the electro magnet (30mA) then?
Any idea how long the 9V battery will last?
 
Yes, you would want to add switch debounce at the 4013 input. Yes, if you want to drive a electro magnet with a 30mA coil then you would use a transistor. The electro magnet coil will draw more current. The current draw of the CD4013 is negligible. How long the battery will last depends on the type 9 volt battery and the amp hour rating based on the battery chemistry. Additionally you need to know the actual total load on the battery.

I looked at your drawings. What exactly is the goal? Why divide the pulses by two?

Ron
 
the 2:1 divider is to correct the reading on the speedometer.
Any tips on the debounce, a 555 maybe?
 
Attached is what you want. On the far left is a debounce example. The switch would go between the resistor cap junction and V+ and the clock in would come from that same junction. V1 in the drawing just simulated 4 Hz as it would come from the switch. Use a CD4013 data sheet for the exact pinout and make sure unused inputs from the unused 1/2 of the chip are tied to common (ground),

Ron
 
I'm still not clear why you need a second reed switch. Can't you just feed the output of the divide-by-two circuit directly to the speedo?
 
Thanks for the reply guys, sorry for the delay I was 1 week at the Costa del Sol:eek:.
@alec_t: The idea is to add this circuit to the setup without altering the original so it is easy reversable. That's why I think of using 1 extra reed as new pickup and the e-magnet as new trigger.

@Reloadron: thanks for the diagram, I wasn't aware of this LTSpice app, interesting to known.
About the clock pulse: I think I need a short pulse in the red diagram, just like the green but divided 2:1. This short pulse needs to drive the electro magnet and I suppose when it will be ON for the duration of 1 complete green pulse (1/4sec here in the example) it will not work. The orginal reed sensor is expecting a short pulse on it is reedcontact from the magnet on the turning wheel. Is a short 2:1 output pulse also possible with the CD4013?
I'm kinda confused with the V1, I understand the idea of it, but in practice when it is not there, where will the CLR pin be connected to then?
Also the output to drive the small electro magnet (30mA), is it possible to complete the diagram a little further? With a transistor?
Sorry for all the questions my electronics are a little rusty after 25 years.
 
The idea is to add this circuit to the setup without altering the original
You will have to alter the original setup to prevent the original reed switch being triggered at the original rate :confused:
 
Yes that's correct, I will move the original magnet on the wheel a bit , away from the original reed and in front of the new one.
So my circuit will then be 'in between' and is fully and easy reversable. That should work I think...
 
Here's a circuit suggestion, using half (U1a) of a CD4013 to provide a debounce function for the new reed switch and the other half (U1b) to divide-by-two the pulses from the first half. The output of U1b is applied via capacitor C3 to drive transistor Q1, and hence your coil, with pulses of ~20ms width, which I'm guessing would be enough to actuate the old reed switch. Much will depend on the effectiveness of the coil. Reed switch coils are normally cylindrical, with the reed inside; but that won't be the geometry you will be using.
The circuit draws virtually no current when the new reed switch is not pulsing it, so no off-switch is needed.
Battery life, as Reloadron pointed out, will depend on active current draw. If the new reed switch pulses at 5Hz and the coil is pulsed at 30mA for 20ms then battery life for a PP3 would be ~ 500mAh / (30mA x 20ms / 400ms) = 330 hours.
 
That's awsome, I'm gonna try it out.
If I understand it alright, the value of cap C3 is depending how long the coil wil be powered? Will a bigger Cap like 4µ7, give pulses then of about 40ms?
What is the function of D1 and D2?
What voltage should C1 & C2 be?
The .asc file what is it? An Autodesk 3D studio file?
Thank you very much, now I'm lot further in my project.
 
the value of cap C3 is depending how long the coil wil be powered? Will a bigger Cap like 4µ7, give pulses then of about 40ms?
Correct. But longer pulses mean shorter battery life, of course.
What is the function of D1 and D2?
D1 allows the cap to discharge between pulses. D2 suppresses voltage spikes caused by the coil's inductance at switch-off.
What voltage should C1 & C2 be?
These are plastic film type, so voltage is not critical. >=12V will do.
The .asc file what is it?
It's a file for running a simulation of the circuit in LTSpice (a useful free download from Linear Technology. This particular sim requires the CD4000.lib model library file available from the Yahoo LTSpice user group).
 
Last edited:
Okay, I can start with that.
I was concidering also of using the smaller 12V (LR23 50mAh) battery instead of the 9V, but that will bring battery live down to about 33h.
What is the best way of adding a small signaling LED in this circuit to see that the 2:1 pulse is working?
Is a small tantalum elco 35V 2µ2 OK to use here?
Does the .arc files also contain the actual drawing? Can LTspice make a veroboard layout or normal circuitboard layout from this circuit?
thanks again.
 
I was concidering also of using the smaller 12V (LR23 50mAh) battery instead of the 9V
It may have too high an internal resistance to provide 30mA +.
What is the best way of adding a small signaling LED
See the attached mod to the circuit.
View attachment 65524
Is a small tantalum elco 35V 2µ2 OK to use here?
Yes.
Does the .arc files also contain the actual drawing?
It shows what the gif image shows, plus extras used for simulation only.
Can LTspice make a veroboard layout or normal circuitboard layout from this circuit?
No. Eagle Lite can do a schematic and pcb layout, and Pebble can help with doing a veroboard layout.
 
Last edited:
great,
C4 is there for making the led burn longer than the 20ms or am I wrong? EDIT: 470µF it's kinda big, no?
What is the use of D3?
1/4W resistors are okay to use I suppose?
While selecting the components I start wondering why not to use smaller battery? Like 2xAAA (3V) = upto 1500mAh, while a 9V Lithium = 1000mAh. The CD4013 can work in this range and I saw also 3V (buzzer) coils too, but what the rest of the circuit? Does it need some changes then?

Any tips for selecting the coil (the electro magnet for activating the small reed)? I was looking at https://www.conrad.com and other places but I didn't find any small electro magnets, so my choice was to use the coil of an magnetic buzzer.
EDIT: I did some testing now with the coil and read switch. I used a 12V coil from a small relay (AZ530-09-2) as electro magnet but the reed is not closing.
EDIT2: yes it does, see post 16/7
If I compare this with the small permanent magnet, this permanent has lot more power, it picks up a paperclip from about 10mm distance, the coil just can hold the paperclip when I hold it against.
So the coil from a buzzer will not do I suppose, any tips for an electro magnet with comparable power like a small (diam 5mm) permanent magnet? In the range 3-9V, max 15mm high, width can be more, 30+. I'm start wondering will the current of the coil still be in range for this circuit/battery?

Thanks for the fast replies, that is a real motivation for me.
 
Last edited:
C4 is there for making the led burn longer than the 20ms or am I wrong? EDIT: 470µF it's kinda big, no?
What is the use of D3?
C4 and D3 enable the LED to stay lit (i.e. > 20ms) when the coil pulse ends. Even with 470uF there will be brightness changes at the coil pulse rate. Reducing C4 will make those changes more noticeable as the LED will turn off between pulses.
1/4W resistors are okay to use I suppose?
Yes.
why not to use smaller battery? Like 2xAAA (3V)
The circuit would be happy at 3V with one or two resistor value changes perhaps (I'll investigate). Any problem is likely to be in energising the coil enough to operate the old reed switch, as you're already finding. Do you have space to wind a few hundred turns of, say, ~32-40 gauge enamelled copper wire around the reed (because ideally the reed is inside the coil) ?
 
Can the LED work w/o D3 & C4? It will be very short pulse then but I think will be visible, no?
The Q- on U1a is not connected in the circuit? I see a small line?

The old reed is fully encapseled in a plastic waterproof housing, I don't see a solution in winding the old reed. Besides it's then no longer possible to easy remove the circuit again.
Maybe there is a solution with a coil that centers it magnetic field all at the top or 1 point, if that exists? The old reed with its magnet may has upto 15mm space for a good working. Of course the electro magnet is not turning with the wheel, so it can be mounted w/o any space.
However I have no expertise in these electro magnets, so I not able to find something useable. Can you advise?
About the component list: I had a look in my old stock, I found BC547B (instead of the 2N3904) and 1N4007 (instead of 1N4148) is that OK?
But before start building the circuit, I need to be sure that I can trigger the old reed with a coil tough.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top