Nominal voltage leaking through MOSFET

[Steveinherts]

Just wired a new unregulated box mod using irlb3034pbf n channel MOSFET + 15 k resistor, checked the 510 connector with a multi meter and it's showing 0.39 volts with the master switch off, is the MOSFET defective or is it ok, wondered if the multi meter might be drawing power and opening the gate slightly, or do I need a higher value resistor.

i am all but ready to pva my box together, thought I'd run it by you guys before as changing components after will be a ....h.
the image is how I've wired it up except I have wired a volt meter independantly with its own on off switch, I have also used 4 x 18650 each with their own fuse, I don't intend vaping below 0.6 ohm so build is for batt life rather than creating extra cloud cover in the sky.

 

[Katmar]

Let's bump you up and see if an expert can help you out. Good luck...

 

[Bunnykiller]

are you isolating the volt meter completely with a double pole single throw switch?? if not, there could be some current supplying the gate somewhere....

would be nice if you had a schematic of the switch and meter shown....

 

[Steveinherts]

I have a rocker switch controlling the volt meter, the negative wire goes to source ground where the MOSFET source comes from and the positive runs through the 2pole rocker feeding off from the positive between the fuses and the 510, I did not tack into the gate supply wire with the master and fire switches, thought the volt meter was independent from the MOSFET circuit that way.

I'm not in a position to produce a diagram as at work at mo but everything else is as the diagram shows, I take it there should be zero volts getting through so is it possible soldering has screwed the MOSFET up, I used 14awg main feed and 18awg for the fire circuit.

 

[kartoffelfaust]

Where did you get the mosfet from? Mouser or amazon/ebay? The ones on the latter sites can be cheap crap, and will melt under load (or not, you roll the dice).

 

[skoony]

your getting drainage leakage between the drain and source.
see spec sheet posted below.

http://www.irf.com/product-info/datasheets/data/irlb3034pbf.pdf

mike

 

[Steveinherts]

I got the fet from stealth vape uk, I looked at the spec sheet ??? So is there a short/drain that shouldn't be there or is it just something I have to live with? Sorry but I've never used fets before and have no idea beyond re-creating a proven design.

 

[skoony]

I got the fet from stealth vape uk, I looked at the spec sheet ??? So is there a short/drain that shouldn't be there or is it just something I have to live with? Sorry but I've never used fets before and have no idea beyond re-creating a proven design.

every semi conductor has leakage current but being so
low it is not a problem.
what is the purpose of the mosfet?
as i read the schematic you have battery voltage
minus the voltage drop of the coil at the drain.
your measuring the voltage difference between the drain
and the source caused by leakage.
the coil resistance is minimal as compared the resistance between
the drain and source in the off condition and drops most of the voltage.
regards
mike

 

[Steveinherts]

The MOSFET is to handle amps larger than my switches are capable of dealing with - right?
The voltage measured are at the 510 + & - points WITHOUT an atty fitted, so no load/drain on the circuit, I would have thought there would be zero volts present as the device is idle come switched off with effectively an open circuit but the multi meter clearly measures 0.39v.
Ideally i would look at placing a switch on the negative wire but I have not the space or switch rated at 30amps to do so as a work around.
How much of a drain on the battery's overnight would it be - substantial or negligible ? If I left the atty on the device
Sorry if I'm coming over as a noob. Cheers Steve

 

[skoony]

The MOSFET is to handle amps larger than my switches are capable of dealing with - right?
The voltage measured are at the 510 + & - points WITHOUT an atty fitted, so no load/drain on the circuit, I would have thought there would be zero volts present as the device is idle come switched off with effectively an open circuit but the multi meter clearly measures 0.39v.
Ideally i would look at placing a switch on the negative wire but I have not the space or switch rated at 30amps to do so as a work around.
How much of a drain on the battery's overnight would it be - substantial or negligible ? If I left the atty on the device
Sorry if I'm coming over as a noob. Cheers Steve

i finally figured out the amp and switch part.
disconnect the wire from the drain and remeasure from the
wire to the source. also reverse leads between drain and the source.
i suspect you meter is loading the circuit some how.
mike

 

[Steveinherts]

Ok will try that when I get home later this evening, my multi meter is only a cheap £15 job but I don't have any other device, would the volt meter work to measure across the 510 instead or would that also load the circuit.

 

[TechnicalEcstasy]

In that circuit the fire switch just pulls current through the resistor. That current creates a voltage drop across the resistor (at the gate of the MOSFET) and turns it on.

1) With the fire switch open you should measure no voltage across the resistor. Verify that with the voltmeter .

2) If there's no voltage across the resistor the only thing it can be is leakage current. Basically a MOSFET is not like a switch exactly. It's more like a resistor. When it's on it has very low resistance and lets current flow. When it's off it has very high resistance and stops all but a small amount of current flow (leakage). It's never really "open" like a switch. Leakage is normal. However, whenever current flows through a resistor it creates voltage so some voltage is normal source to drain.

Ohms law dictates that .39 volts across a .6 ohm coil should consume 650 milliamperes. That should kill a 650 milliamp-hour battery in of course one hour. Doesn't sound right. I don't work with MOSFETs often so I don't know offhand what is.

Like skoony said, sounds like something is loading the MOSFET Down.

 

[Steveinherts]

Thanks for all your input, this a head scratcher, check, no voltage across the resistor but pos on batt > neg on drain pin of fet is now showing 0.63v.
The only 2 things not mentioned is the charge circuit, but that is wired to the batt source + & - (not plugged in to usb).
the antivandal fire switch has a led ring, neg to source and pos fed from the master switch so it illuminates when pressed.
its just odd why i have this small voltage on the atty posts.
what do i do? say ....... it and glue it up, head in sand or try a new fet or start again with an ugly big a*** 30a switch (def not what i want)
its driving me nuts, i can see no obvious fault.

 

[kartoffelfaust]

Post a picture of what you're working on - may help.

Try isolating various aspects of the circuit - something is not right, whether its wiring or the mosfet.

 

[Steveinherts]

this is the exact wiring of the mod, double checked and then again, all wires are as diagram. there's a lot of wires in a small space but no bare wires touching.
i have removed the fire switch led pos wire but made no difference.
out of curiosity, is someone out there with an unregulated mosfet mod, do you have a meter reading across the atty connector while switched off (without the atty attached), just wondering if i have discovered something about them previously unchecked.
anyhow have a look, see if i've screwed up somewhere, thank you for your patients.

 

[Bunnykiller]

just a thought.... unhook the button light wire from the gate and test the 510

 

[TechnicalEcstasy]

Hmmm. I did a circuit emulation and the emulator is predicting that you would measure the full 3.6V (battery voltage) where you are measuring. I inserted a resistance of .5 ohms to simulate a heater coil and it started conducting without the fire button even being pressed. I would have predicted the circuit should work fine. Something odd is going on for sure. Have to see if I can find an engineer to run it by tomorrow.

Is this a proven circuit? Have other people used it successfully? If they did then did they use the same MOSFET?

 

[Steveinherts]

i removed the led from the gate and made no difference.
the design was from find my vapes and i believe it is the same mosfet, think its what most people use, i did a lot of research before buying my components.

 

[TechnicalEcstasy]

Ok. I ran this circuit by an electrical engineer this morning and he confirmed that it should work in theory. He also confirmed what I initially expected, that there should be no significant amount of voltage measured at the drain of the MOSFET with the fire switch open. It seems that you’ve done everything right so far and measuring the unexpected voltage was a good catch. The circuit itself could be improved with a couple of more resistors but it should work ok as is. The engineer suspected the same that’s been discussed here, that you have a leaky MOSFET. Everything so far points to that.

 

[Steveinherts]

Thank you for your time and trouble, I have picked up a new 3034 fet today, maybe the heat damaged it, will try that first, I have a 100ohm resistor on the switch led already as it's rated 2volt, what ohm and where could other resistors be of benefit? If I still suffer from leakage I am thinking of making a special circuit breaker switch to handle 30 amps to place between the batts and the source pin so the device is totally isolated, the cut out on my wood panel is only small so look like it could be a challenge on my lathe and milling machine, this would be my last ditch work around, where there's a will there's a way.

UPDATE - the new mosfet was a disaster, it was leaking nearly 2 volts, changed the resistor to a new one - no change, still nearly 2 v so put old fet back in, the old fet is showing 0.4v.

Will now design a switch to isolate the fet from source, need it to fit small cut out as want to keep symmetry on the box, the volt meter switch will be used for the gate and the volt meter will now be wired to come on with the fire button.

the mod objective was to provide lots of power with 12,000 mah (4 x 18650 panasonic greens 3400mah) for extreme run time between charging and to stop taking the batts out to do so.
The box is already made, with the faffing around with the electronics this seems the easiest work around as i don’t want the batts draining overnight, it will also give me a double safety value against accidental firing. UNLESS someone can shed some light on a fix to prevent leakage, i think it's the next best solution bar throwing the fet away.
i will post the results and pic's if it's successful.

Thank you to all who have contributed to trying to solve my dilemma,
steve.