Need a way to make this circuit touch sensitive

Hi to all,

I am trying to make a lighter that was broken work again.

It used to be a touch sensitive one and that part of the circuit died. However I have found another old lighter that has a similar battery powered sparking mechanism but leaked, so I am cannibalising that one for parts.

The only thing is that the second lighter was sparked by a switch, a standard sub microswitch in fact. and using all 3 contacts.

I have made a tiny touch senstive circuit using a darlington transistor and a couple of resistors but I can't figure out how to integrate it into the lighters spark circuit since I only have the grounding effect of the darlington to play with.

I need to make something which will emulate the 3 contacts of the microswitch but take very little room. The whole circuit is dead until opening the lighter closes a separate switch from the battery so there is no need to worry about any drain when not opened.

Any ideas would be much appreciated, but remember I need to keep this small.

Thanks Al

PS I have posted my poor effort at drawing the circuit in "paint", I know its a bit rubbish but I don't have anything else to draw it out with.

The good news is you can just put a 1k resistor from 1 to 3 and then just use an SPST (pushbutton) switch from 1 to 2. It will work the same, and so long as you don't hold the switch down for any great length of time (minutes) it will not affect battery life. Try it and see!

The bad news is your darlington probably won't cut it. This circuit needs a very brief, high rise-time, high-current pulse through that ignitor coil. Darlingtons never "saturate", the best they can manage is about a .8V drop, and you aren't even likely to get that considering the current requirements for any coil that can generate a spark. Also, a resistive touch switch does not have anything like a fast rise-time. What this means is that, as the darlington s-l-o-w-l-y starts to switch "on", current will be bleeding out of that cap through the coil, reducing it's voltage. In the few milliseconds (or less) it takes to come full "on" the voltage will be probably be too low to trigger the spark.

Try it anyway. Boost the voltage with another battery, see if that helps. Ultimately I think you are going to need a slightly more complicated circuit with a mosfet to drive the coil.

Hi Duffy,

I can see what you mean about the rise time. The trouble is that I need to keep the circuit small. There is a small cube shaped place in the lighter about 15mm x 15mm x 9mm. there is also a touch panel built into the lighter body so I don't need to account for that.

I have never used either spark circuits or touch sensors before and thought it would be simpler! Doh!

I do have PIC chips in my bitsbox so I could use say a 12F629 to do the job, but I am unsure how I could interface the PIC to both parts of the circuit. Do you have any ideas? You are obviously knowledgeable about spark circuits and such so have a ponder mate.

EDIT. As the original lighters spark was made by a microswitch I wondered how fast the contacs of the switch were in relation to the speed of the darlington turning on fully?

Thanks Al

OK, I just wrote out a massive reply, and accidentally hit 'home', and lost it all....

Firstly, mechanical switches switch pretty damn fast - from almost infinite resistance, to very low resistance very quickly. In this circuit, it seems at 'idle' the supply charges up a capacitor... once full the cap has massive resistance, so really doesn't drain the power at all. When the switch is pressed, the capacitor suddenly discharges through the coil, (which is probably a camera flash trigger transformer?). Any resistance in the path of the switch slows down the discharge, giving much less voltage on the output = little, if any, spark. A single transistor if somehow turned on instantly, still might not be able to switch the high current spike fast enough.

So you need something that can take a rather large amount of current, for a very short period of time as the cap dumps all its energy across the coil - mechanical switches/relays are ideal. But solid state should work Either a power transistor (PNP), or a Pchannel MOSFET.

Some idea's:
As suggested, darlingtons are probably too slow, and would reduce the voltage from the cap to the coil (higher saturation voltage than a single transistor). So, maybe drive the main transistor with a JFET? - those are often used in touch sensitive circuits. Or perhaps a logic IC. CMOS logic (CD4000 series) would be ideal as they can run at 12V, where-as a 74HC series needs 5v, and so would require a regulator (as would a PIC micro-controller ...which is over kill!)

I'm sure a single transistor is capable or switching the cap (1 to 2 on your schem), but that transistor needs to be saturated quickly.
As it switches from a charged cap (+12V) to the load (the coil) a PNP should be used. To turn that on, its base needs to be pulled from 12V, to <11V via a resistor. An NPN (or N channel MOSFET) could be used for this, although it might not have enough gain to turn the PNP on fast/hard enough for it to switch that current. - maybe an NPN darlington for this. Thats a total of three transistors (darlingtons can be in one package, so thats only two devices). Again, and FET could be used, but like the CMOS logic would need input protection.

This page has loads:
http://talkingelectronics.com/projects/TouchSwitch/TouchSwitch-1.html

I'll admit, I've never seen a lighter sparker that complicated, its mostly piezoelectric, or flint.... or at the other end of the spectrum, for oil burners, 25-50kV

Is this a hand held thing? cigarette lighter? with a small 12V battery? if thats the case, then I'm sure a power MOSFET is overkill - not to mention too big, as would a power transistor. So many options! If you've got a pic of what you're dealing with, space requirements, it should narrow down whats easiest/best. I still say, best case, couple of transistors - worst case, a transistor and a small logic chip

Sorry duffy, see you've already mentioned a MOSFET as the switch :/ I feel like a right tit now.

Hi Guys,

First off here are some pics of my lighter. This was my favourite reliable everyday lighter until it got washed by accident! I did dry it and try it but it was dead and when I took it apart things got even worse, with me breaking one of the tiny coils in it removing the innards.

Anyway I really would like to repair it as it was so nice to use and lit every time. The lighter I used as a donor was just a cheap battery lighter that had failed with a leak, the spark still working ok, but as I said earlier this one worked by a microswitch so I need a touch sensor circuit to replace the switch. The underside of the microswitch can be seen connected to the red coil unit from the donor. The coin for anyone outside the UK is a 2p which are exactly an inch in diameter (handy as a measure sometimes).

The unit is only powered when the lighters lid is lifted so power usage is not too critical.

Anyway onto what I found! While looking for a suitable method I came across the circuit below that works from a 555 and saw a very similar one on the site Duffy sent. It made me wonder if I could power a suitable FET with the output of the 555?

Also for some reason I have loads of N channel FETs but not a single P channel so if it is possible with the 555 I would need to use a N channel, or it means waiting for a delivery (and finding more parts I don't really need) to justify ordering some.

So if anyone has an idea that will work please let me know.

Thanks again.......Al

Hi Guys,

Just been pondering on something. Would the circuit posted below work?

I have tried to rearrange things so I could use an N channel FET.

Any comments?

Al

Edit. Would a 2N7000 FET be enough? Says its rated at 200mA 60v. 500mA pulsed!

No, a 2N7000 has a 5Ω "on" state resistance, that's going to eat your power.

Something like this little thing -
IRLML6244TRPBF
- has an Rds of only 0.021Ω, and can handle a brief 32A pulse. Also, it has a gate threshold of less than a volt, is smaller than a grain of rice, and is cheap.

A 555 has a nice sharp risetime. If you're going to use one of those, try the CMOS version - like the ICM7555IPA - because these have much, much lower operating current. it will essentially be "on" the whole time the lighter's lid is open, and a regular 555 can draw up to 7ma, which is significant on those tiny batteries. This CMOS version only draws .03ma.

Looks like theres not quite enough space for a relay and 555, and as already pointed out the 555 is permanently powered, so it'll draw current, draining the battery. Also, with a relay that will draw significant power when switched, so ideally you need it to switch on for a fraction of a second.

Here's what I've come up with, trying to make it as simple as possible, using the parts you have (but not the 2n7000!). The use of a PNP darlington is a bit dodgy here.... because it'll b e turned 'on' when its base is below ~10.8V... and if its floating... it probably will be. So perhaps instead of two PNP's in a darlington, use a PNP at the top, switched by an NPN at the bottom. The input to the NPN is your touch plate. But hey, this is just for sh*ts n giggles.

Operation:
Lid is opened connecting MOSFET's source to ground (batt-). User touches plate, which turns on the upper PNP, and switching the MOSFET's gate 'high' via a capacitor. The cap is there so the MOSFET is only switched on for a brief period. (otherwise, constantly touching the plate would shunt current through the mosfet and 1K resistor, to ground, which means roughly 12mA drain whilst holding).

The 330uF cap gets charged up via your coil, and a 1K resistor regardless of whether the lid is open or not. When the lid is closed (mosfet not connected to GND) touching the plate shouldn't turn the mosfet on. When the lid is open (connecting MOSFETs source to GND) and the user touches the plate, the mosfet switches, it essentially connects the coil across the cap, which dumps its load into it.

As you can see from the graph, big current spike! (this is an 'ideal' simulation, for proof of concept, I guessed the inductance of your coil, and its all a bit 'ish

Edit! I re-read your post several times and just realised you meant the batteries negative terminal is only connected to the circuit when the lid is open. so.... that means in my schem you can forget about the 'lid switch goes here' and just connect all the grounds together - to the lid switch - which then connects to the batt -. Simpler! As for a suitable MOSFET, I simulated some common ones, most worked fine (literally like using a switch), but you'll want one with a low RDSon (which means rated for high current, 6A+) and prefereably small. A surface mount SOIC8 mosfet will do just fine, and you can solder wires direct to the pins as theres only three connections (4 pins for drain, 3 pins for source, and 1 pin for gate). All that will easily! fit into the space.

One question: Is the casing of the lighter gorunded? as in, connected to the battery's negative terminal?

This is probably the simplest form of a Darlington array touch switch. It simply uses two transistors and a couple of resistors. It may work for your lighter.

**broken link removed**

Der Strom

P.S. I don't condone the use of cigarettes. I hope this lighter is more for show