hello

i'm trying to use a MOSFET as a switch is this possible.???

the problem i have is that i need it (one) in the way that none of the 2 wires sides are conected to ground...!!

soow what i need is a mosfet whit 4 pinss..????

one side for the low power (DC) and the other side for the wire it has to connect.. hope you can help me...

i also need that the resistenca is near by 0.. when rds on..

The amps i need to conect are 8 Amps.. (+/- 12VDC)

TKS

 

Hi TKS,
A regular N channel MOSFET will allow current to flow from a positive potential to a less positive potential if the gate has a higher voltage than the source. (some will argue I know ).

The source does not have to be at ground potential for a FET to work.

If you could tell us more about your circuit we could help you more.

Thanks.........Tony

__________________
I know I learned everything at least once,
now where did I write it down?

 

Do you need fast switching? If not, use a relay.

 

There are plenty of FETs out there that meet your needs. You are looking for a power mosfet. So your turn on voltage will be between 2-4V.

Ground is just a reference point, nothing special about it. As long as you can bring the gate voltage up to 4V with respect to the source, the drain to source channel will form and you will have low RDS on. How low? Pick a mosfet that fits your need (look at the datasheet). Here is a hint about that, if the datasheet says RDSON is 0.5 Ohms, that is true only under the conditions that they show they tested it for. If your application does not put it in those exact conditions(unlikely), your RDSon for that part may be higher or lower by some amount.

Here is another hint, you need your "switch" to run 8 amps? Do not choose a fet based on ID continuous current max = 8 AMPS! That device can handle 8 amps but probably for only very low voltages. What you need to do is calculate your power dissipation in the device, pick a fet based on that and make sure it can handle greater than 8 amps. Whatever device you pick, will only survive if it is operating within its SOA curve. This curve includes voltage across the fet, current through the fet and power dissipation.

I have seen to many people fall prey to this. They wire it up, turn it on and it dies even though they never exceeded volts or current rating. Ah, but they exceeded power rating and the die became so hot, it passes the glass transition temperature and the SiO2 melts. If you have this part on a heatsink, it may not even feel hot! There could be a huge temp gradient between die and heat sink.

The message is, do your homework on the operating conditions before you pick a device.

Fairchild semiconductor has nice lines of power fets. You can even get many of them sampled to you for free.

If you are not sure about how to wire the drive up for the device without using "ground", show us your schematic.

Remember, a device doesnt know what "ground" is - only you do and it is inapropriately named to boot! I assure you, my "ground" here is not at the same potential as yours.

 

to draw you a schemtic is a little use less because in words you could imagine the same.

the thing i want to switch is the negative side of a car bobine ''coil"

soow the + is conected to 12volts...and the negative is conected to different point and one of them is the ignition module.
when there can flow from the coil to the ignition module
some 4amps at highest..then the engine can run.

Whe are building a rpm limitter sow we want to cut of that cable.
nothing more nothing less (yes i know all the goods and bads about it..)

i have bought some BUZ11 es and i'm testing whit them.

the problem i had where that when i gave the gate a puls to conect them
the engine could run but when i turned of the pulse the engine kept on running....

in the image you see the sheme..

the hard thing is the thing i didn't finis.
because to switch whit 5volts i need a lover reference right..??

today i discoveret that i also need to ground the gate lin whit a resistance to the 0 line because else it keeps hanging...

well hope some body gives me the finishing touch

TKS

p.s. when conecting the bobine to ground directly there run 16amps.!!
and the engine don't run..!!!

p.s. maybe whit some high resistances between 0 and the mod line..???14K for reference.??

Attached Thumbnails

 

 

For safe working i recommend: drive the FET via optocoupler.

Attached Thumbnails

 

 

after burned 2 mosfets...

it still don't work.

thirst i tried it with your sheme
but then the mosfet keeps the switch closed
and then it doesn't work anymore. :evil:

2nd i tried the gate to 0 with a 50K resistence but than
it worked (slowly) with my multimeter (bleep pos.) :twisted: :lol: :shock:

but when i connect it to the coil. it still don't work. When i supply the power, the engine runs; but when i cut of the power it keeps running.
(mosfet kapoet..!!! :lol: )


soow hope somebody can help me..

TKS

 

Do your mosfets have a diode between drain and source like in the drawing above? if it doesn't then the EMP of the coil when you turn it off may be what shorts your mosfet.

Place a diode yourself or get mosfets wich can handle inductive loads.

 

Yes,

there is a diode inegrated between Source and Drain (in that way)

S--->|-- D when i measure it it says 0,535 volt needed for the treshold..

i don't know nothing more, could you explain a bit more..??
They told me also something about a shotky diode on the coil..

From 0v -->|--12v but i don't know..

TKS

 

i also have nformation that the tjob of the module
is to put the ground line of the coil to 0v. and then realising it
whil producing the spark.

the funcion of the module seems to bee that. soow my mosfet is never for the fully 100% conected to ground..

actually it is grounded 300 times per second (@ 6000Rpm...)

and 6*500 / 60 = 50 times per second (@ 1000Rpm)

is that the problem...??

do i have to put a extra diode..?? in serie..???

TKS

 

Tell us the part number of the fet so we can go look at the datasheet.
I'm not at all shocked that your fets are blowing.

 

earilier wrote: i have bought some BUZ11 es and i'm testing whit them.

the problem i had where that when i gave the gate a puls to conect them
the engine could run but when i turned of the pulse the engine kept on running....

in the image you see the sheme..

the hard thing is the thing i didn't finis.
because to switch whit 5volts i need a lover reference right..??


sow its a BUZ11 (i have today brought the last 3 on stock..!! :P )

p.s. could upload it soww sorry..

TKS

 

The place where you have inserted the FET is going to have BIG voltage spikes. a kilovolt would be no surprise. If the gate drive is referenced to ground, the FET will be destroyed at the first spark. Using a FET in the positive lead of the coil would work if you could arrange that the FET is only switched off when there is no current in the coil. This is one of the few instances where I am going to recommend a relay! A capacitor of about 0.22uf 1000v connected across the contacts will reduce sparking.

 

no boy,

you didn't get any chit out of the picture...!!

the ground kabel from the coil is the way that the spark module switches on the coil..


soow lets start where it begans thats by the rotor there is a small "hall gever" ( a kind of token giver) then the module conects the ground of the coil to ground and then come the spark in no time up..

soow the kabel we are switching is not thespark cable but the ground of the coil wich switches very often to ground per sec.

sow it is 6/12 volts.. ( the coil is in seri whit a 1ohm resistence (coil also has about 1ohm)

but about 4 to 6amps..!!!



TKS

 

Just an idea: if the wires of hall-element accessible, You can blocked it direct with opto.