mosfet as switch

If you can't show the circuit then please try to describe its connections.. My guess is that you're using it incorrectly!.

This Mosfet has a body diode in it, which means if you apply + voltage on the source, it'll be there at the drain. And the Gate is a voltage control device not a current device as you are using in terms of (BASE). You should at least have a pull down R to ground so that you keep the gate discharged!. IT's a very high Z input which can maintain a charge. You also can damage it.

Since this component is a N channel device, you need to pull the (-) terminal of the battery to common via the Drain connection of the NMos and the Source goes to common..

The Gate will be switched on via what ever voltage needed. I didn't look at all of the spec's

the circuit is really simple. connection the ground of two batteries, the positives are not attached to anything. yes I could use an alligator clip and hook up the two grounds together. But I would like to use a semiconductor. I would need some sort of mosfet but when the gate is activated, it can let current flow in both direction, is there anything on the market that can do this?

I think it is called a UJT, just need to find one in pspice...

Are you saying you want to put these batteries in series ?

I think at this point you maybe better off using a relay..

two grounds together, that would be parallel, But the UJT is taking too much current just to enable it. The problem is that when I put a MOsfet with source on one grounds and the drain on the other ground, the batteries are in parallel. no matter what the gate voltage is...

Remember Occam's razor.

Don't use a transistor when a switch will do; don't use an IC when a transistor will do........

Batteries connected like that will do nothing, more details would help.

MOSFETs will do that they just have that annoying body diode that makes them always conduct (like a diode) in the backwards direction.

you may be able to escape this problem by connecting two back-to back.

connect the gates together and the drains, then connect the sources to the batteries and the gates to your control signal.

Bye. Jasen

"lerameur"

UJT = what ?

Uni - junction transistor ???

The term you want is " BJT " - f****it.

..... Phil

What I want to do is to be able to fully control by battery bank at a distance. I have some batteries that are fully charged and others need to be charged. I would like to switch over a specific battery from one battery bank to another without moving it. So, in one of the case, I would want to hook up an extra battery in parallel when in the battery bank, therefore I need to hook up the positive to the bank positive and the negative to the bank negative. I am having a bit of problem hooking up two identical voltage.

Trying not to get blown up by his lead-acid battery bank? ;-)

Michael

I posted the circuit on the following link:

The idea is to disable M1 so only battery V2 V3 and V4 are used.

I'm not sure I fully understand what your trying to do,but if your switching between batteries you should use a bidirectional switch. This stops body diode conduction of the FET.

I dont know your load but here is a pretty good PFET if it's beefy enough for you.

You can get ten at Digikey for a little over Ten bucks.You will also need a mosfet driver.

Here some reading for mosfets. 100 page PDF from fairchild .

Made a mistake in my drawing. Those two mosfets should be connected source to source.

The problem with that design is the internal diode that parallels each MOSFET. For one direction of current, each MOSFET will conduct through that diode, even if the gate is reverse biased. Even if you reconfigure this parallel battery arrangement to have each battery negative connect to a single common node node (instead of a ladder structure) you still have to account for both battery charge and discharge current (if the batteries will be charged in place). That requires a pair of MOSFETs in series, an N-channel and a P-channel, so that each one blocks for the other when its internal diode is forward biased.

The you have to get the gate biasing right.

I have been reading this pdf from above

I have been reading this pdf from above

I am the one who made a mistake. You need two MOSFETs, so that you can point a body diode each way, so that one or the other blocks current, regardless of the current direction. They can both be N-channel, both P-channel, or one P and one N-channel, as long as the body diodes are pointing both ways. The choice of type depends on which way the gate biasing is most convenient.

That the current can go both ways is exactly the problem. You get conduction through the channel (in either direction) in parallel with the body diode, but one or the other MOSFET must have a body diode reverse biased if the pair is to be an open circuit when the MOSFETs are both off.