Low Voltage Cut Out

Status
Not open for further replies.

CraigHB

Ultra Member
ECF Veteran
Jul 31, 2010
1,249
314
Reno, Nevada
This is as far as I would go for extra protection.

2920L300DR Littelfuse PTC Resettable Fuses

Very low additional resistance. 3A hold, and 5A trip sounds about right.

I'm using the 1812L260THDR Littlefuse PTC part in my current mod. I run two of them in parallel to double the trip current and cut the resistance in half. I'm getting 18mΩ with the both of them, or 18mV per amp. Really happy with that. I was using a single Bourns PTC fuse before with similar trip current as the two, but it had 60mΩ, big improvement.
 

mamu

Ultra Member
ECF Veteran
Verified Member
Mar 29, 2009
2,551
1,523
IN USA
...For example: one RC site suggests a 3.9v zener, an hi-output led, a fixed resistor and a trim pot to calibrate it. Oddly, the zener is put in backwards (I don't understand it exactly). Then just hook it up to a variable ps and calibrate the cutoff where you want. Light dims as it approaches cutoff, goes out completely when time to replace batts.

That would work, but the problem is with the dimming of the LED. Using a shunt regulator (has a built in reference voltage from a zener diode) also does the job. The nice thing is there is no dimming of the LED - the LED is either on or off.

In some of my mods I added a LM431 shunt regulator as a batt status monitor. If the LED lights, batts' voltage is greater than 6.5v. If the LED doesn't light, the batts' voltage is less than 6.5v. At least it lets you know when it's time to charge the batts.

Here's how I wired it in the mod (I used a separate small tactile switch for it, but I guess you could also wire it to the atty switch):

lm431.jpg


For 6.5 volt indicator:
R1 = 3.3K ohms
R2 = 2K ohms
R3 = 500 ohms
R4 = 500 ohms

Changing R1 will change the volt value. For example, if you wanted a 7.0v indicator, use a 3.6K ohm resistor for R1.
 

mamu

Ultra Member
ECF Veteran
Verified Member
Mar 29, 2009
2,551
1,523
IN USA
P.S. Thanks for the tip Cape. I'm going to use your suggestion. The mod I'm working with has very little room for extra parts. Definitely not enough room to add the shunt regulator circuit. Adding a zener and resistor shouldn't take up too much room.

ETA: Going to order the parts now. From what you're indicating, a 6.5v zener + 680K ohm resistor would shut the regulator off when the batt's voltage is at 6.5v. Is that correct? Is there an equation you used to calculate the value of the resistor?
 
Last edited:

executivul

Senior Member
ECF Veteran
Mar 5, 2011
76
55
45
Romania
Hi all,
This is how it works: you put a reverse biased zenner and a resistor from Vdd (positive) to Vss (negative) leads of the batteries. By reverse biased I mean the zenner with the cathode to the positive side. A zenner diode will allow current to pass in the normal biased mode (cathode to the negative) OR in reverse biased mode (cathode to the positive) IF the voltage drop is greater than the zenner's specified voltage. The resistor in series with the zenner is put to the negative side, and is value is large enough so it won't draw a lot of current (10k ohm or more). How that translates in non engieer talk? If you have less than zenner's voltage the zenner will be closed, that means all the voltage drop will be on the zenner itself and no voltage drop on the resistor, so between the negative lead and the solderpoint zenner-resistor you will have 0 voltage. If you have more than zenner's voltage the zenner will start conducting current, the voltage drop on the resistor will be the difference between battery voltage and zenner's voltage. Current through the resistor is very small so it won't dicharge your batteries, you just use the solderpoint between the zenner and the resistor as a voltage reference.
Hope I made myself clear since English is not my primary language.
 

CapeCAD

Senior Member
ECF Veteran
Verified Member
Jun 2, 2010
232
25
Massachusetts
P.S. Thanks for the tip Cape. I'm going to use your suggestion. The mod I'm working with has very little room for extra parts. Definitely not enough room to add the shunt regulator circuit. Adding a zener and resistor shouldn't take up too much room.

ETA: Going to order the parts now. From what you're indicating, a 6.5v zener + 680K ohm resistor would shut the regulator off when the batt's voltage is at 6.5v. Is that correct? Is there an equation you used to calculate the value of the resistor?

The value of the resistor is not critical, it is there to connect the pin to gnd when the voltage to the zener is less than the breakdown voltage, and yet high enough resistance to prevent a short to ground when the voltage is greater than the breakdown voltage. A smaller value would use more current, and a larger value would use less but take longer to shut off.

Resistance is calculated the same way you would calculate an atty. A 680k resistor will use approximately .01milliamps when converter is operational (I=V/R), similar to using an LED.
 

WillyB

Vaping Master
ECF Veteran
Oct 21, 2009
3,709
591
USA
I'm using the 1812L260THDR Littlefuse PTC part in my current mod. I run two of them in parallel to double the trip current and cut the resistance in half. I'm getting 18mΩ with the both of them, or 18mV per amp. Really happy with that. I was using a single Bourns PTC fuse before with similar trip current as the two, but it had 60mΩ, big improvement.
Interesting, but why "double the trip current". 5A seemed just about right.
 

CraigHB

Ultra Member
ECF Veteran
Jul 31, 2010
1,249
314
Reno, Nevada
Mainly to lower the resistance. That one trips in 2.5 seconds at 5A, but will still trip at 3A over time. It measures out at 36mΩ. I would have rather used a single one with a 4 amp hold and a resistance under 25mΩ, but couldn't find anything like that. It seems the actual resistance is always in the middle of the range they give. The last one I used was the same way. It's okay though, my battery can put out 10A all day and the other circuity can handle the surge current as well. I've done repeated short tests on it and it's fine. Normally, there's a voltage detector on the output and the µcontroller shuts down the output in the event of a short in no more than 200µs anyway. The PTC fuse is just my hail mary backup. Can never be too safe with these batteries, especially the big ones that hold lots of energy.
 

MadmanMacguyver

Ultra Member
ECF Veteran
Feb 8, 2011
1,473
607
Dallas Texas
Mainly to lower the resistance. That one trips in 2.5 seconds at 5A, but will still trip at 3A over time. It measures out at 36mΩ. I would have rather used a single one with a 4 amp hold and a resistance under 25mΩ, but couldn't find anything like that. It seems the actual resistance is always in the middle of the range they give. The last one I used was the same way. It's okay though, my battery can put out 10A all day and the other circuity can handle the surge current as well. I've done repeated short tests on it and it's fine. Normally, there's a voltage detector on the output and the µcontroller shuts down the output in the event of a short in no more than 200µs anyway. The PTC fuse is just my hail mary backup. Can never be too safe with these batteries, especially the big ones that hold lots of energy.

Uggh and thats why I'm taking some EE courses...not for a cert but to have the material needed set down in front of me in a straight forward manner...I know some and half understand some but Its a needed thing as mods get more and more complicated...

and I am using pre-made protection boards...
 

bstedh

Ultra Member
ECF Veteran
Verified Member
Apr 5, 2011
1,551
581
Northern Minnesota
www.clear.com
Yeah, you need to do some study to really understand all this stuff. I'm sure a basic understanding can go a long way.

At least we can still go super simple =] Battery / Switch / Connector... The rest of it is luxury add on and safety. Plus the fun of tinkering =]
 

CraigHB

Ultra Member
ECF Veteran
Jul 31, 2010
1,249
314
Reno, Nevada
True, you can take it to whatever level you want. You can have zero understanding of electricity/electronics just to do a basic box mod with a battery and switch.

I do this stuff for the challenge and accomplishment so I go all the way with it on the circuit design end of things. Perfect for me since I happen to be an electrical engineer and a smoker. I was into it as soon as I discovered vaping. I think I started thinking about my first mod the first time I took a hit off an e-cig.
 

MadmanMacguyver

Ultra Member
ECF Veteran
Feb 8, 2011
1,473
607
Dallas Texas
well actually I can go far beyond that...I have designed 10 diff pcbs and dremel etched one and am gathering supplies to chem etch my next one...I can easily understand a pre built device and generally go from there as to how to put it together w the passives to make it work...I can usually handle warping a device to serve my purposes within its tolerances and sometimes extend those tolerances to the bare edge...I have not figured out how to build a circuit from passives to do what I want beyond simple 3-6 component devices and that takes me a LOOOOONG while...Hence needing EE courses...I want...no NEED to know more...
oh dang gonna have to get another appointment w gummy soon...:lol:

Nice reg by the way Craig HB you still using them or did you ever...

I think I started thinking about my first mod the first time I took a hit off an e-cig.

Me too...I started tiny and graduated to maniacal in no time flat...now I'm adding so many features I have to have learn to program microcontrollers too...grr learning curve is too steep here...
and I would love an actual cutoff for undervoltage...only thing I have now is relay based...unless I buy premade...
 
Last edited:

CraigHB

Ultra Member
ECF Veteran
Jul 31, 2010
1,249
314
Reno, Nevada
Nice reg by the way Craig HB you still using them or did you ever...

I use the TPS43000 in my latest mod. My last one used the MAX1709. The TI part is a really nice controller, very efficient. I get peak efficiency of 98% and minimum efficiency of 92% at 30W. It averages around 96% at the wattage I typically use. Using it in boost mode with a single battery. The down side of that chip is the external compensation network it requires. That took me quite a while to get dialed-in. Compensating that chip is a full on exercise in electrical engineering. It's all analog stuff, yuck, poles, zeros, bode plots. Out of my comfort zone there. I've always been into the digital end of things.

I plan to keep using the TPS43000 until I find something I like better. I've been tempted to have a go with the Linear Technology LTC3785. It's a really trick buck/boost controller that switches between buck and boost mode by self-configuring with 4 FETs. It's a really novel and ingenious approach to a difficult task. Highly efficient as well. It uses more parts, but all these controllers have their pluses and minuses.

Me too...I started tiny and graduated to maniacal in no time flat...

Maniacal is a good way to put it. I've been spending pretty much all my free time modding since I started. Though I just finished my latest mod and I'm taking a well deserved break for a while, browsing the forums and enjoying the new device. I've already got a couple new projects waiting the wings. I think of this stuff way faster than I can build it.
 

mamu

Ultra Member
ECF Veteran
Verified Member
Mar 29, 2009
2,551
1,523
IN USA
The value of the resistor is not critical, it is there to connect the pin to gnd when the voltage to the zener is less than the breakdown voltage, and yet high enough resistance to prevent a short to ground when the voltage is greater than the breakdown voltage. A smaller value would use more current, and a larger value would use less but take longer to shut off.

Resistance is calculated the same way you would calculate an atty. A 680k resistor will use approximately .01milliamps when converter is operational (I=V/R), similar to using an LED.

Thanks Cape.

Can this setup (zener + resistor) be used in line with a P-MOSFET? With one leg of the atty switch wired to the MOSFET's gate and the other leg of the atty switch wired to the neg. batt terminal, how would the zener be wired?

I'm thinking of wiring the anode end of the zener to the control pin of the module which then the cathode end of the zener is wired to the FET's gate and from there wired to the atty switch. But I'm wondering if this will create a problem what with the other leg of the atty switch wired to the neg. batt terminal or if it will interfere with how the MOSFET works or maybe create a short with the 680K resistor wired to ground.

I really would like to add this safety feature to using these switching modules so it will shut off if the voltage gets below 7v (digikey didn't have 6.5v zeners and the next closest was 6.8v).

It would be something like this...
fet.jpg
 
Last edited:

bstedh

Ultra Member
ECF Veteran
Verified Member
Apr 5, 2011
1,551
581
Northern Minnesota
www.clear.com
With the control pin you don't need the P-Fet. It is doing the same thing as the control pin on the regulator. The control pin is going to have low current draw so there is no need for it. And it may cause unintended issues with the internal control pin circuit.

Also purely from a design stand point I prefer to put the main power switch on the positive battery terminal. This is not really necessary in a plastic enclosure but if anyone uses a metal enclosure with your design they may run into issues.

I will let others chime in on the placement of the zener but at first glance I think you have the right idea.
 
Last edited:

mamu

Ultra Member
ECF Veteran
Verified Member
Mar 29, 2009
2,551
1,523
IN USA
With the control pin you don't need the P-Fet. It is doing the same thing as the control pin on the regulator. The control pin is going to have low current draw so there is no need for it. And it may cause unintended issues with the internal control pin circuit.

Also purely from a design stand point I prefer to put the main power switch on the positive battery terminal. This is not really necessary in a plastic enclosure but if anyone uses a metal enclosure with your design they may run into issues.

I will let others chime in on the placement of the zener but at first glance I think you have the right idea.

Thanks bstedh! That's good news that I don't need a 3A switch if using the control pin. The switch I plan to use in this mod is a NO switch rated at 100mA so I thought I needed the FET.

The enclosure I'm using is plastic and it's a really tight fit for what I want this mod to have, so not having to wire in the FET definitely makes my day!
 

CapeCAD

Senior Member
ECF Veteran
Verified Member
Jun 2, 2010
232
25
Massachusetts
Thanks bstedh! That's good news that I don't need a 3A switch if using the control pin. The switch I plan to use in this mod is a NO switch rated at 100mA so I thought I needed the FET.

The enclosure I'm using is plastic and it's a really tight fit for what I want this mod to have, so not having to wire in the FET definitely makes my day!

Hi Mamu,

I believe what you have will not work, you are connecting the on/off control pin to ground through your atty switch when it really wants to be connected to +Vin. As bstedh pointed out, the P-FET is not needed.

Now, you have several options...I haven't tried these, speaking in theory only.
1. Do not use the zener and resistor, but use your detection circuit output (LM341-C) connected to the on/off control pin to control the uvp. If the LED lights, it will enable the converter. The cathode current of the LM431 should be sufficient to drive the enable pin (1mA) as long as you have it set to > +2V to light the LED. I would still use a 100k resistor from the on/off control pin to ground if you do this. Atty switch can be to ground as you have it. This would allow you to adjust UVP to what you want.
2. Do not use your detection circuit and use the zener diode and resistor, then add a resistor and LED to the on/off control pin. If the on/off control pin is high enough to work, the LED will light. Atty switch will be from on/off control pin to +Vin.

Either way, you do not need as many components and both should perform the same function. Your detection circuit will allow you to adjust the uvp level.
 
Last edited:

mamu

Ultra Member
ECF Veteran
Verified Member
Mar 29, 2009
2,551
1,523
IN USA
Hi Mamu,

I believe what you have will not work, you are connecting the on/off control pin to ground through your atty switch when it really wants to be connected to +Vin. As bstedh pointed out, the P-FET is not needed.

Now, you have several options...I haven't tried these, speaking in theory only.
1. Do not use the zener and resistor, but use your detection circuit output (LM341-C) connected to the on/off control pin to control the uvp. If the LED lights, it will enable the converter. The cathode current of the LM431 should be sufficient to drive the enable pin (1mA) as long as you have it set to > +2V to light the LED. I would still use a 100k resistor from the on/off control pin to ground if you do this. Atty switch can be to ground as you have it. This would allow you to adjust UVP to what you want.
2. Do not use your detection circuit and use the zener diode and resistor, then add a resistor and LED to the on/off control pin. If the on/off control pin is high enough to work, the LED will light. Atty switch will be from on/off control pin to +Vin.

Either way, you do not need as many components and both should perform the same function. Your detection circuit will allow you to adjust the uvp level.

Thank you so much Cape!! Great suggestions! Using either option is a great resolution to a safety concern and also give me an LED status indicator. I'm leaning toward option 2 as it would take up the least amount of space.

Thanks again. Very much appreciated!
 
Status
Not open for further replies.

Users who are viewing this thread