Another mosfet/reed question

[mastou]

I want to try to build a dive light for spearfishing and I would like to use a reed/mosfet switch. I thought of using 3xXR-E or 1xP7 running of 3-4 D-size NiMh since I can get more time in the water this way instead of using AA-size (and also because I have some great D-sice NiMh's) I just need a good boost driver then if using 3xXR-E in series

However, I am no electric expert and could therefore really use some feedback on my choice of reed contact and mosfet.

Mosfet:
http://elektronik-lavpris.dk/files/sup2/irl530npbf.pdf
N-LogL 100V 17A 79W 0,10R TO220AB
Vgs+/-16V
Vgsth min.1V max.2V

Reed contact: http://elektronik-lavpris.dk/files/sup2/ksk1a66.pdf
SPST 0,5A 10W AW10-20
Glass piece is only 14mm so pretty small

Finally, I have read that I should use a 10K resistor between the gate and source of the mosfet would this do the trick: http://elektronik-lavpris.dk/product_info.php?products_id=83327 or would it be better to use a metalfilm resistor? and why exactly 10K?

Sorry if my questions are stupid - my excuse is that I am still to build my first light

/Magnus

 

[VidPro]

10k is a lot of resistance, it is probably for pulling the Mosfet gate back down.
the gate itself doesnt require a lot of current, and the current it does require is quick and happens to push the gate when the mosfet switch goes ON (so to speak)
so
i am saying it is likly that the resistance applied in this situation is just for the transister gate movement, and has little requirement for large current capability. (but robust and overdone never hurt nothing)

once you trigger the gate on the mosfet with your switch thing, it seems to LOCK into place (All loose terms) so you have to bring the gate back down after you remove the power used to switch it.
so a switch (like reed) can turn it on, by applying power to the gate, once the power is removed from the gate, you still have to push the gate back down to get the mosfet switch to go off again.

one big concideration when using a mosfet as a switching item, is to get it to open and close FULLY, which is important when you have less total voltage in the curcuit (like 3x1.2v batts).
say we have a 4V gate trigger voltage (4v required to switch fully) and only apply 3.7v to that gate because of low battery. the mosfet will not switch fully and instead have a lot of resistance.
As Long As
the mosfet has a low resistance when closed , way less than a ohm even, AND it is closed fully , then it can handle much and much amps.
if it becomes a resister in there, then it can not handle the amps they are rated for the same way at all.

so they got tons of specs, because (like a lot of stuff) if operated Exactally right, even a small one is capable of large switching. When things dont work as planned, then a big fat well heat synced one still survives the heat.
so make sure you got a good mosfet gate switching, even at lower battery power, and everything goes fine. Or put a big fat one in, and everything always goes fine

i had NO idea what the heck was going on when i started playing with mosfets, as Relays were solid as a rock and worked everytime, and dont ever cause parasitic drain paths and all.
i would still use a relay for some stuff given the space and all.
and old school transisters were rated for any situation they could be in, at any time. mosfets MAX rating is only for when they are switched Fast, and Closed (and opened) Fully.

here is "my First Mosfet" reads just like a childrens book http://www.candlepowerforums.com/vb/showthread.php?t=84242 if it was written by one
it has Pictures.

the thing that isnt in there, is the voltage Dividing. the Gate wants a certian voltage , and potentially you can burn out a gate with high voltages. they use resisters to divide the voltage you have, to hit the gate with the correct voltage. THEN the battery goes down , and it again doesnt trigger fully . With the low voltages your talking, frying gate shouldnt be an issue.

i now (for some reason which i dont even understand) use HexFets, which can still be FAT, have very low restance, and will trigger fully even with lower voltages.

 

[VidPro]

http://elektronik-lavpris.dk/files/sup2/irl530npbf.pdf
now i need EE help, cause i cant read these sheets, i do hands on.

the FET you linked is the nice package, that can be synced and handle being half open/closed, but its "On resistance" looks way higher than the new stuff with less resistance.

then
like Evan said in that thread the "a low maximum Vgs voltage you have to watch out for" (for frying the gate)
but like i am saying you want full gate trigger, so with low voltages you need the "lower" voltage gate triggers, like 1.8v and 2.5v . WITH that voltage it is fully on at the lowest resistance, not just close, still having .3ohms.

i thought Vgs was the gate to source voltage, or the "gate voltage" simply. the one you linked to shows a high vgs thing. it wants 4V just to get going good, and handles 16V
so it doesnt LOOK like a good choice for the lower voltages??

 

[HKJ]

However, I am no electric expert and could therefore really use some feedback on my choice of reed contact and mosfet.

It does look ok.
The resistor is not very critical and there are no reason to buy a more expensive one.
The power mos can not be used for 17A current, but will work fine with the current required for the led you specified.

I would add a 100 ohm resistor in series with the read contact, again you can just use a cheap resistor.

 

[Steve K]

I want to try to build a dive light for spearfishing and I would like to use a reed/mosfet switch. I thought of using 3xXR-E or 1xP7 running of 3-4 D-size NiMh since I can get more time in the water this way instead of using AA-size (and also because I have some great D-sice NiMh's) I just need a good boost driver then if using 3xXR-E in series

However, I am no electric expert and could therefore really use some feedback on my choice of reed contact and mosfet.

Mosfet:
http://elektronik-lavpris.dk/files/sup2/irl530npbf.pdf
N-LogL 100V 17A 79W 0,10R TO220AB
Vgs+/-16V
Vgsth min.1V max.2V

Reed contact: http://elektronik-lavpris.dk/files/sup2/ksk1a66.pdf
SPST 0,5A 10W AW10-20
Glass piece is only 14mm so pretty small

Finally, I have read that I should use a 10K resistor between the gate and source of the mosfet would this do the trick: http://elektronik-lavpris.dk/product_info.php?products_id=83327 or would it be better to use a metalfilm resistor? and why exactly 10K?

Sorry if my questions are stupid - my excuse is that I am still to build my first light

/Magnus

hi Magnus,

for a battery powered light, I think that this mosfet is not the best choice. I think that a mosfet rated for 20v or so would work better. It would allow the mosfet to be smaller, have a lower resistance (Rds), and would usually be cheaper.

The mosfet should also be a "logic level" mosfet. This means that it will only need a few volts applied at the gate to be fully turned on. I think the mosfet you suggested needed at least 4v to be fully on. This might be good enough, depending on the light.

You should be able to find a good mosfet is a much smaller package. It might be as small as a SOIC-8, or a SOT-23. I'd recommend these, but you might not be good at soldering something so small, so maybe a medium size package like a DPAK would be better? The IRLR3717 could be a good choice:
http://www.irf.com/product-info/datasheets/data/irlr3717.pdf

I think you are on the right track, but look for a mosfet with a Vdss rating of 20v or so. A rating of 100v is much, much higher than you need. Look for a logic level mosfet too, which will be easier to find in a mosfet with a Vdss of 20v or so.

good luck,
Steve K.

 

[mastou]

Thanks for all the answers

@ VidPro thanks for the link I have already read your thread there was alot of great info.


Ii it the maximum the maximum or minimum Vgs that is required to turn the gate on fully? and would I be better of using a fet with higher max Vgs instead of using a resistor (like you suggested HKJ)?

There are plenty of other fets to choose from, see list under "Technical Specifications": http://elektronik-lavpris.dk/product_info.php?products_id=77985

Thanks for all the help

/Magnus

 

[VidPro]

the minimum for full closure is important to me, when it goes to its lowest resistance.
then the maximum voltage it can TAKE is important to keep from damaging the part permenentally.

the max voltage is often at least 2x over the min for full closure. so i usually get enough leeway for battery high and battery low.

 

[HKJ]

Thanks for all the answers

@ VidPro thanks for the link I have already read your thread there was alot of great info.


Ii it the maximum the maximum or minimum Vgs that is required to turn the gate on fully? and would I be better of using a fet with higher max Vgs instead of using a resistor (like you suggested HKJ)?

/Magnus

The mos you had selected in you first post was already a logic level.
You need to look at the graph called "Typical Output Characteristics", find the current you are using and be sure your Vgs are high enough to be on the horizontal part of the curve, even with a nearly depleted battery.

The extra resistor I suggested was not due to the mos, but to secure a long life for the reed contact. It has a maximum current of 0.5A and without the resistor it will have a over current each time the light is turned on.

As long as the maximum battery voltage is below the maximum Vgs, you do not have to worry about high Vgs voltage.

 

[mastou]

@HKJ - I totally forgot to look at the reed contacts data...would it be better to use a bigger reed contact since I would be running 0.7-1amp (if using 3xXR-E in series - parallel seems to be a bad idea from what I have read) or 2.8amp (if using a P7).

Is it correct that a NiMH has a maximum voltage of 1.4 (fresh of the charger) and a minimum voltage of 1 (when recharging is needed)? Because since I was thinking of using 3-4 D-size NiMH maybe I should choose a fet with a Vgs(minimum) of 3V and Vgs(maximum) of 6V.

I know it would probably be better if I just decided to use either 3 or 4 NiMH but I would prefer if I could build a switch to use in both cases.

Again thanks for your patience with a newbie

 

[HKJ]

@HKJ - I totally forgot to look at the reed contacts data...would it be better to use a bigger reed contact since I would be running 0.7-1amp (if using 3xXR-E in series - parallel seems to be a bad idea from what I have read) or 2.8amp (if using a P7).

Is it correct that a NiMH has a maximum voltage of 1.4 (fresh of the charger) and a minimum voltage of 1 (when recharging is needed)? Because since I was thinking of using 3-4 D-size NiMH maybe I should choose a fet with a Vgs(minimum) of 3V and Vgs(maximum) of 6V.

I know it would probably be better if I just decided to use either 3 or 4 NiMH but I would prefer if I could build a switch to use in both cases.

Again thanks for your patience with a newbie

As long as you uses a mos to drive the led and only need the reed contact to open the mos, any small reed tube will do.

You have to look at the curves to see how the mos performs with you voltage and current, not min. and max. Vgs specifications. With the IRL530 you need 4 volt to get it open, then it will have a voltage drop of about 1 volt at 9 A (At 3A it will only drop 0.3 volt).
The IRLR3717 is a much better mos for switching, at 3.5 volt Vgs it will drop less then 0.1 volt at 10A. At 3 volt it can handle 4A with 0.1 volt drop.

NiMH can get a little above 1.4 volt fresh from the charger and 1 volt is a good point to charge.