Z41 Smart Switch

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
Hi all :wave:

Introducing the SmartSwitch - a high power, low resistance, dual momentary flashlight clicky switch.

Physical Details

There are 3 versions of the smart switch:

Z41 ULP Switch

The Z41 ULP SmartSwitch is an ultra-low-profile switch that once installed consumes less than 7mm of battery space (boot and spring included). This is ideal for use in setups that require maximum utilization of the tube space available. Due to the physical design of this switch it is only compatible with Z41 tail caps or similar.​

Z52 Switch

The Z52 SmartSwitch is a direct replacement for a McClicky switch and shares the same physical dimensions and threading properties.​

Z57 Switch

The Z57 SmartSwitch is compatible only with SureFire Z57 clicky tail caps.​


Electrical Details


While the physical packaging of the 3 switches are different, the functionality and electronics of the switches themselves are identical. The details are as follows:


  • MOSFET based switch
  • Forward and reverse momentary operation
  • Microprocessor controlled
  • CR1025 3V Coin cell powered
  • MOSFET rated to 20V and 30A absolute maximum*
  • Coin cell voltage monitoring
  • In standby mode the switch will last over 10yrs.
  • In active mode the switch will last the equivalent of leaving a light on for 3 days straight.

The 3 day switching time is based on the user actively holding the switch in. If the user has permanently turned the switch on and then released the switch (so the light is still on) the MCU returns to a deep sleep mode. If you had a main battery large enough you could turn this switch on once and it would run for over 10 years. Simply put, under normal operating conditions 3 days is the absolute worst case scenario, with typical use far exceeding this time.

Testing completed at this point in time (2.5 months of daily use) has revealed that the coin cell is yet to drop below 3V. Despite having been in use for 2 months, this cell is effectively brand new.

* I strongly recommend that the MOSFET ratings are not pushed too hard, with safe usage theoretically being around 10V/15A.


Theory of Operation


People seemed to like VanIsle's FETtie switch. In principle it was a great idea. However I saw a few reported issues with it and after doing a bit of research it boils down to the following.

FETs have a turn on Voltage (Vgs​) which when reached means the FET is fully on. At this fully on state the resistance from the Drain to Source (Rds​ - basically the resistance between the flashlight host and the battery negative terminal) will be at its lowest. This value differs from FET to FET. MOSFETs will start switching on even if Vgs ​is not reached. At this lower on-voltage Rds​ is increased. With increased resistance comes an increase in power being burnt off as heat. FETs can only handle so much power before they go up in smoke. The FET used by the FETtie, whilst capable of turning on properly with voltages as low as 3V, does not actually turn on properly much lower than that. So while the FETtie would have worked perfectly on a fresh cell, not long after first use it would have slowly started cooking itself. The more the coin cell is allowed to deplete, the worse the situation gets.

Eventually every FETtie would fry itself unless the coin cell was replaced reasonably regularly.

So the key was to find a FET that can turn on properly at sufficiently low voltages for this application. The FET the SmartSwitch uses guarantees full turn on at 1.8V. In order to ensure the switch doesn't attempt to operate at too low a voltage, a microcontroller has been incorporated to monitor the voltage of the coin cell. Once a low voltage threshold is reached the SmartSwitch will dim and return to full brightness 0.5s after it's turned on, and then once every 2s. This warning is only present while the switch is actively pressed. To allow for any errors in the accuracy of the voltage readings, the low voltage cut off in the SmartSwitch is 1.9V.

There is a secondary advantage to having a microcontroller/MOSFET pairing inside the switch - we can have forward AND reverse momentary action. No more picking one or the other :thumbsup:

The switch is basically the same as used in digital cameras. The momentary action (opposite of the current permanent state) is enacted with a half press and a permanent action is enacted with a full press. It's totally different to a McClicky as there is a very distinct difference between the momentary and permanent action. Operation of momentary mode is near silent, with a permanent press being audible, but nowhere near as loud as the click of a conventional switch. The best example of this type of switching would be the pre-focus on a digitial camera.

Proof of Concept and Functional Demonstration

Lsfedfe.jpg




Z52 Render

pWNFpXI.jpg


Z57 Render


f12UoSa.jpg


As always any questions, comments, and suggestions are encouraged.

Thanks,

- Matt
 
Last edited:

climberkid

Flashlight Enthusiast
Joined
Jan 19, 2008
Messages
2,100
Location
Goldsboro, NC
Talk to me more about the lockout options for this.

I love the idea of this switch.
Is it running the hyperion programming or am I missing the intent of this switch replacing or not replacing a multi mode driver?

Sent from my SAMSUNG-SGH-I337 using Tapatalk
 

Tofty

Enlightened
Joined
Dec 28, 2011
Messages
276
Location
Hampshire, UK
Excellent switch Matt.
Sounds like, electronically speaking, the switch is pretty much ready to go, which is always nice to hear, just some minor mechanical and programming considerations to address.
Looking forward to seeing this switch for sale.

For the low voltage warning; will it trigger every time you try and change modes or just when the light is turned on and off? if it's only when the light is switched off and on then perhaps the switch could cycle through modes every half a second for 5 seconds or so then not blink again allowing the light to be used without distraction. The mode cycling should very clearly catch the users attention but then not immobilise the light thereafter. A mode change every two seconds would leave the light pretty much unuseable which would be a real problem in an emergency. I first though it would be a good idea to have the switch cycle through all the light's modes once or twice then leave you with the mode it would have started on originally, but i realise that is impossible as the switch cannot know how many modes the light has.
 

Moddoo

Flashlight Enthusiast
Joined
Nov 19, 2008
Messages
1,243
Location
Minnesota
Looks good Matt,
When we spoke about this earlier this year, I was also prototyping with a 2 stage switch.
I have chosen to take my setup a slightly different direction. It is taking some time to perfect with other things eating up a lot of my time.
Looks like you are pretty close to finishing this up. I will be interested in testing it out when available.
My general policy is to not discuss my projects in public until they are ready to deliver. I'll not say more until that is true.

Best of luck on this, it seems to be working well.
Cheers
 

mvyrmnd

Flashlight Enthusiast
Joined
Sep 4, 2009
Messages
3,391
Location
Australia
Very very cool.

Will this fit in an E-series tailcap, as a McClicky without the brass ring does?
 

archimedes

Flashaholic
Joined
Nov 12, 2010
Messages
15,780
Location
CONUS, top left
Is this just for high-current handling, or are there any additional features (such as soft-start, pwm modes, voltage blink-out, strobe, etc) ?
 
Last edited:

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
Is it running the hyperion programming or am I missing the intent of this switch replacing or not replacing a multi mode driver?

It is not running any "Driver programming". It is intended to compliment existing multi-mode drivers. There a re a few reasons for this but the main one is that in order for this to "control" a driver/light/whatever, said light would have to be single mode only. I also don't know how well the FET I've chosen would react to PWM switching. I might have to dig around though it is certainly not my intention for this capability to be included. It could possibly be done, but then it'd be a fundamentally different device to what I posted above.

Excellent switch Matt.
Sounds like, electronically speaking, the switch is pretty much ready to go, which is always nice to hear, just some minor mechanical and programming considerations to address.
Looking forward to seeing this switch for sale.

For the low voltage warning; will it trigger every time you try and change modes or just when the light is turned on and off? if it's only when the light is switched off and on then perhaps the switch could cycle through modes every half a second for 5 seconds or so then not blink again allowing the light to be used without distraction. The mode cycling should very clearly catch the users attention but then not immobilise the light thereafter. A mode change every two seconds would leave the light pretty much unuseable which would be a real problem in an emergency. I first though it would be a good idea to have the switch cycle through all the light's modes once or twice then leave you with the mode it would have started on originally, but i realise that is impossible as the switch cannot know how many modes the light has.

Electronically speaking yes. The prototype above is "ghetto'd together" lol. I mean it won't change much, but I am waiting on some special components to arrive to make assembly cleaner and easier. There's also the small issue of a battery carrier that we previously spoke about. I'm getting dimensions as we speak ;)

The voltage monitoring is activated only when the button is actively being pressed. So the low voltage blinks will only do their thing if you are holding the switch at either position. If you have it on permanently (and are thus not holding the switch) it won't give any indication that somethings up. That's why I currently have it set to 0.5s - it's fast enough that you likely won't miss the blink if you're using the momentary feature.

You're spot on about the number of modes being an issue in whatever host the switch is installed.

I'm actually thinking climberkid might have struck a cord earlier. What if PWM dimming did work? If it was fast enough would the light even know it's being turned on or off? In this case could I possibly set the low voltage warning to actually be a "dip" in output? Press the switch and the light quickly (like over the space of 2 seconds or something) ramps down and then back up. No mode change, and it won't do it again until you release and press again.

Might have to look into it.

Looks good Matt,
When we spoke about this earlier this year, I was also prototyping with a 2 stage switch.
I have chosen to take my setup a slightly different direction. It is taking some time to perfect with other things eating up a lot of my time.
Looks like you are pretty close to finishing this up. I will be interested in testing it out when available.
My general policy is to not discuss my projects in public until they are ready to deliver. I'll not say more until that is true.

Sorry if I've inadvertently forced you to reveal your cards. It was not my intention :( PM sent.

Is this just for high-current handling, or are there any additional features (such as soft-start, pwm modes, voltage blink-out, strobe, etc) ?

As sort of described above anything other than switching the lights main power supply is tricky without changing the entire focus of the switch. My goal with this switch was to develop a better FETtie. I think I've done that. The dual momentary functionality was a bonus.

Soft start functionality might be doable if the PWM option works. I don't really know much about soft-start but I believe this is focused more on incans? Are there any LED lights that require soft start? How would an LED driver react to soft start?

It's readin' time!

- Matt
 

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
Will this fit in an E-series tailcap, as a McClicky without the brass ring does?

Totally missed this for some reason. As it currently stands the switch fits the entire inside dimensions of a McClicky adapter. This means that the switch is the same size as the McClicky, including the threading. I'll need to work out how deep the threading goes, but if I can get the switch to fit in an e-series host it would be cool :)

I should point out that the brass ring in the original post is a home made McClicky adapter. So the switch is the PCB parts, and is not inclusive of the brass ring in the pictures.

- Matt
 

moshow9

Flashlight Enthusiast
Joined
Jul 10, 2010
Messages
2,222
Location
El Paso
I must plead ignorance in regards to these types of switches, but I do have a question: Is the CR1025 3V Coin cell user replaceable? Or is it a one and done type thing once the coin cell has been depleted?
 

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
I must plead ignorance in regards to these types of switches, but I do have a question: Is the CR1025 3V Coin cell user replaceable? Or is it a one and done type thing once the coin cell has been depleted?

The coin cell is fully user replaceable. No good having to throw out the whole thing just because the power source is gone. That being said I am taking steps to ensure replacing the cell is as simple as reasonably possible. There's no getting away from having to take the switch out of the light to replace the coin cell, though given the run time of the switch it'd be a once in a blue moon sort of thing.

Also my 3 day rating in the original post is based on the user actively holding the switch in. So if you have permanently turned the switch on and released the switch (light is on) it's not using anywhere near as much current. Basically put, under normal operating conditions that 3 days is severely under rated :)

- Matt
 

ShineOnYouCrazyDiamond

Flashlight Enthusiast
Joined
Aug 18, 2009
Messages
4,442
Location
CT, USA
Matt,

This is a great offering and I would definitely be interested in one once completed.

My concerns are this:
- Switching time. Will this have any issues working with multimode drivers like your Hyperion of other available firmware options for something like a 105C driver? I ask this because I've had a Tri-EDC threaded to take a V10R Ti switch and I had some initial issues with mode changing on a custom firmware 105C driver. The problem was in the movement of the button. With too much movement it took too long to switch on and off and thus would change level consistently. I had to put a spacer inside the switch so that it would take almost no motion and just a little bit of pressure to turn on and off and then things worked properly. Having watched the video it appears you have already addressed this issue and although the initial operation seems a bit confusing it appears that it would work very well.
- Fitment - I think it would be great that this could be used anywhere as a McClicky replacement option.
- Coin cell battery - I am concerned with the 3 days on time, even if extended to say a week with a slightly lower on current. Have you considered using something like a rechargeable LiIon coin cell or a button cell capacitor (like used in the V10R e-switch)? That way it could charge when the light is off and the user wouldn't be required to change button cell batteries? I am just thinking long term frustration for heavy users who only use that one light and use it a lot having to replace a switch battery often.
 

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
Switching time. Will this have any issues working with multimode drivers like your Hyperion of other available firmware options for something like a 105C driver? I ask this because I've had a Tri-EDC threaded to take a V10R Ti switch and I had some initial issues with mode changing on a custom firmware 105C driver. The problem was in the movement of the button. With too much movement it took too long to switch on and off and thus would change level consistently. I had to put a spacer inside the switch so that it would take almost no motion and just a little bit of pressure to turn on and off and then things worked properly. Having watched the video it appears you have already addressed this issue and although the initial operation seems a bit confusing it appears that it would work very well.

Touch wood, zero issues. The Hyperion firmware is pretty reliant on fast button pressing and switching, and in all the testing I've done so far the only mistakes I've made is because I apparently have difficulty counting to 8 :duh2:. The switching is instant. I guess the best way to explain it is that with a McClicky or similar switch, the switch actually "switches" over a range. Once the light activates there is still some room to move before it clicks into place. This switch has 3 very distinct positions and the movement between each position feels instantaneous. You can't really feel the move from one position to another. Stupid but relevant example: it's kind of like looking at a finch's (the bird) head - one second it's looking left, and even though you swore it didn't move it's suddenly looking right. Cute little b*stards.

Fitment - I think it would be great that this could be used anywhere as a McClicky replacement option.

Yeah my goal tonight will to be to check tolerances.

Coin cell battery - I am concerned with the 3 days on time, even if extended to say a week with a slightly lower on current. Have you considered using something like a rechargeable LiIon coin cell or a button cell capacitor (like used in the V10R e-switch)? That way it could charge when the light is off and the user wouldn't be required to change button cell batteries? I am just thinking long term frustration for heavy users who only use that one light and use it a lot having to replace a switch battery often.

I had thought of this a while a go but for some reason dismissed it due to technical reasons. It does depend somewhat on what sort of light engine is being used but so long as there is a mechanical switch where the Smart Switch sits, there is no reason why it would not work. I might have to investigate this. Thanks for the heads up!

- Matt
 

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
Oh quick question. With the V10r e-switch does it come with any instructions, specifically a set of specifications stating it's intended use? Or is it pretty much assumed it would only be used with the V10r and people are just shoehorning it into other lights? I ask because I think I remember why I didn't go down the capacitor path - most suitable caps are rated at 5.5V max. This makes the switch unsuitable for anything other than single cell lights. This removes one of the advantages of the battery in that it's useable voltage range is instantly limited :(

A rechargeable coin cell is also an option but this would require a basic charging circuit and there is next to zero room as it currently stands. That's not say I won't look into it, just that the chances don't seem too good.

I will test my switch thoroughly but I believe you'd be looking at having to swap the coin cell very infrequently. Not as good as never of course.

- Matt
 

psychbeat

Flashlight Enthusiast
Joined
Nov 3, 2009
Messages
2,797
Location
SF norcal
So three days of "momentary" or half press :)
This could mean years of use for those of us that just turn their lights on & off.

I'm excited about this - my FETtie fell apart :(
I've been fine with a McClickie stuffed w copper braid but would def be interested in one of these.

Thanks for the hard work!
 

Mattaus

Flashlight Enthusiast
Joined
Mar 29, 2011
Messages
1,765
Location
Brisbane, Australia
So three days of "momentary" or half press :)
This could mean years of use for those of us that just turn their lights on & off.

I'm excited about this - my FETtie fell apart :(
I've been fine with a McClickie stuffed w copper braid but would def be interested in one of these.

Thanks for the hard work!

Thanks for the kind words :)

I'm literally working on a second revision right now. The planned upgrades are a smaller overall diameter so it can fit into e-series hosts. This would oddly enough allow for a larger battery and what I think is actually a cleaner design!

You have a McClicky? Mind telling me (if capable) what the inner diameter of the threads are? I know a McClicky plus threading is 0.682" in diameter, but in order to fit this into an e-series light I need the diameter without the threads. I had two McClickies but sold one with a host and the other ain't coming out of the light it's stuck in!

- Matt
 
Top