Looking for a NC touch switch, one with no moving parts. I'm getting tired of swapping switches. Anyone have any ideas?
Touch switch?
- Thread starter o4_srt
- Start date
- th_trl_thread_readers 0
- Status
For anyone following this, qt118 (charge-transfer touch sensor) appears to be suitable for use with the evercool circuit. It features a 10 second time-out function, useful for accidental fires, would prevent destruction of an atty/carto.
It uses ground plane (can be made with a piece of copper foil) to sense your finger, then turns on, with vout being vin-0.7
When off, vout is 0.6
The ptr08100w requires -0.3 to 0.6 volts on the inhibit to stay off. It turns on when the inhibit pin sees 2.8 to 6 volts.
If vin for the qt118 is 5 volts, it will output 4.3 volts, which is within the range needed to turn the regulator on.
The only problem is that, from what I understand from the data sheet, the load attached to the qt118 must be able to sink 4 ma at Vol, and the regulator can only handle -125 ua, so I would need to current limit the output of the qt118. (Using a 150 ohm resistor?)
Anyone with a little more theory under their belts care to verify/correct my logic?
Sent from my MSM using Tapatalk
It uses ground plane (can be made with a piece of copper foil) to sense your finger, then turns on, with vout being vin-0.7
When off, vout is 0.6
The ptr08100w requires -0.3 to 0.6 volts on the inhibit to stay off. It turns on when the inhibit pin sees 2.8 to 6 volts.
If vin for the qt118 is 5 volts, it will output 4.3 volts, which is within the range needed to turn the regulator on.
The only problem is that, from what I understand from the data sheet, the load attached to the qt118 must be able to sink 4 ma at Vol, and the regulator can only handle -125 ua, so I would need to current limit the output of the qt118. (Using a 150 ohm resistor?)
Anyone with a little more theory under their belts care to verify/correct my logic?
Sent from my MSM using Tapatalk
wow no takers?
Talked to a coworker yesterday, the QT118 should be able to be used with nothing more than a cap and a ground plane, with the output going straight to the inhibit pin of the regulator.
I'll end up ordering two in the next week or so and breadboard it, so verify that it will work as intended, then incorporate it into my next mod.
Talked to a coworker yesterday, the QT118 should be able to be used with nothing more than a cap and a ground plane, with the output going straight to the inhibit pin of the regulator.
I'll end up ordering two in the next week or so and breadboard it, so verify that it will work as intended, then incorporate it into my next mod.
Let try this again.... 04_srt, I have never seen or use that chip but I would love to get one to play with. If it does not work you could always just use a MosFet touch on the battery supply. There are a lot of post talking about that, One is this one:
http://www.e-cigarette-forum.com/fo...h-mosfet-transistor-tutorial.html#post1808467
http://www.e-cigarette-forum.com/fo...h-mosfet-transistor-tutorial.html#post1808467
Thanks for the info guys! I have parts coming in for new mods and this would be a cool feature to add. There something "active" about using the push button but no moving parts is a good thing. I'm eager to get more info on the QT118 and check that tutorial!
well, went to order the chip, digikey is now out of stock :-(
looks like I'll have to work on a MOSFET version instead. Thanks for the link blue!
looks like I'll have to work on a MOSFET version instead. Thanks for the link blue!
ive glanced at his thread, haven't (yet) taken the time to scour it for information, thanks!
I stumbled upon a new version of the original chip:
AM42QT1010
the max on duration has been raised to 60 seconds, but I think with a 555 timer, could be reduced to under 10 seconds to avoid carto/atty damage.
I tested the evercool tonight, with the outputs of the mentioned chip, the PTR08100W functions exactly as if it had a mechanical switch. Will be ordering this new part in the near future, once I get a parts list together. Digikey has them for roughly 86 cents each.
The advantage of using this chip over a MOSFET switch is that there is no contact necessary. The sensor can be placed inside the mod, and can detect through the case. It is not dependent on moist fingers. Additionally, it can act as a proximity sensor, which, with some testing, may yield a more reliable automatic battery pack. Combined with a master sensor to sense hand placement on the mod, and a proximity sensor, it could be the ultimate automatic mod.
now to find a suitable watertight, ruggedized case....
AM42QT1010
the max on duration has been raised to 60 seconds, but I think with a 555 timer, could be reduced to under 10 seconds to avoid carto/atty damage.
I tested the evercool tonight, with the outputs of the mentioned chip, the PTR08100W functions exactly as if it had a mechanical switch. Will be ordering this new part in the near future, once I get a parts list together. Digikey has them for roughly 86 cents each.
The advantage of using this chip over a MOSFET switch is that there is no contact necessary. The sensor can be placed inside the mod, and can detect through the case. It is not dependent on moist fingers. Additionally, it can act as a proximity sensor, which, with some testing, may yield a more reliable automatic battery pack. Combined with a master sensor to sense hand placement on the mod, and a proximity sensor, it could be the ultimate automatic mod.
now to find a suitable watertight, ruggedized case....
Hey! That's cool! Thanks for the update! Here's the link to it at digikey. They have an "Evaluation kit and demonstration board for AT42QT1010." (at Atmel) but it sounds like you might not need all that....
The app notes are pretty thick...
You got the chip and tested it already and it worked??? You're quick!
What did you use for caps? Is the sensor on top of the switch? I'd love to see some pics when you get going! I have various parts coming in but am not sure what to do whit them yet.......
Last edited:
There have been many posts with that design, including discussions of resistor values and even ready to etch PCB layouts.
did not get the chip yet, I used the rated outputs of the chip (Voh = 2.3 volts, Vol = 0.5 volts) on the inhibit pin of the PTR08100W, and the regulator functioned as intended.
I'll probably order a few different cap values, as the value of Cs affects sensor sensitivity. That way, I can fine-tune it.
Willy, I'm aiming for something different, I'd rather not rely on completing the circuit with my finger, and I aiming for a waterproof battery box. Less holes for water to get in is always better.
Waterproof can be dangerous. No holes means if the batteries go into a thermal meltdown. You basically have a grenade With the atomizer connector being the weakest link. Might not be good if you were puffing on it. LifePO4 batteries would be ok i guess, since they are not supossed to explode like li ions. But unless you use three of them they wont work very well with the PTR08100W.
You can always drill some vent holes and put a waterproof sticker over them. I'm sure if the battery went it would blow right through the sticker.
Good point, and to the response, good solution.
Lifepo batteries could work, I'm not too big on small mods, I like something with some size, they are more comfortable to hold.
Sent from my MSM using Tapatalk
Check this out. It might fit your needs - NC OR NO - Completly sealed and explosion proof for those times the battery goes kawaam 
. Unable to source it yet -- guess one could e-mail them for a list of distributors
. Unable to source it yet -- guess one could e-mail them for a list of distributors
Check this out. It might fit your needs - NC OR NO - Completly sealed and explosion proof for those times the battery goes kawaam
I saw that during my search.
I ordered 5 of the amtel chips today, 85 cents each, along with some sot-23 to DIP breakout boards to make the chips easier to work with.
If I can get it to work, I'm planning a mod using two sensors and an and gate. The mod won't fire unless your hand is in vaping position. Will be more difficult to pull off though, as the voltage supplied by batteries is not constant, so ill need a regulator to provide vcc for both the and gate and the amtel chips.
First build is not going to feature a regulator for the chips, just a voltage divider. If I set vcc for the chip at 5.5 at 12 volts, at 7.2 volts (min voltage supplied by batteries), the chip gets 3.3 volts, which is within vcc specs.
Will update when I make some progress, hopefully they come by the end of the week.
Sent from my MSM using Tapatalk
Check this out. It might fit your needs - NC OR NO - Completly sealed and explosion proof for those times the battery goes kawaam
They work great...BUT are 19mm across. I even tried to custom order them.. $$$$$$$$$ way toooooo much.
That's what I was afraid of.
Got the atmel chips and the ptr08100 in today, waiting on a shipment from mouser, then the build will begin. Hopefully Monday.
I'm also thinking about ditching the inhibit pin, and using a MOSFET as a switch between the cap and regulator, to minimize the 1 mA leakage current. Might also incorporate one between the output and load. This would further increase efficiency by not allowing the caps to discharge upon removal of power.
This would work quite well, i believe, limiting leakage to less than 10 uA
Last edited:
The amtel chip works alone in a breadboard, using an LED as a load. However, I can't get the sensitivity high enough to use as a proximity switch, still have to touch the electrode. Ill have to play around with the cap values and electrode size to tune it to sense through a plastic box.
The good news (other than it works) so far? Moist fingers not required.
Sent from my MSM using Tapatalk
The good news (other than it works) so far? Moist fingers not required.
Sent from my MSM using Tapatalk
- Status
Similar threads
