Hi,

I have some basic knowledge of electronics, but I'm having trouble trying
to get some real design to work. One of the things that do not work as
expected is the BS170 MOSFET I'm trying to use. I know, it's not a very
sophisticated part, but I have a few of them lying around and I thought
they would be able to assist my learning.

Anyway, I connect the gate and source of the BS170 both to the ground,
and the drain to a 330 ohm resistor, connected to 3.6 V. When I then
measure the potential of the drain, it's at 2.85 V, which means it draws
a current of (3.6-2.85)/330 = 2.3 mA. Thus, the drain-source resistance
is about 1.2 kohm. My expectation was a much larger 'off' resistance (say
100 kohm or more).

I cannot relate anything on the data sheet (http://www.fairchildsemi.com/
ds/BS%2FBS170.pdf) to my observation. I tried with another BS170, which
gave about the same results. Is my expectation wrong? Apparently, it is,
but what is the right explanation? And, how can I deduce this from the
data sheet? Then, how to solve this? Should I just use another
transistor, or is there some common way to handle this?

In my case, I connected a LED to the drain and it didn't shut off. I can,
of course, put a resistor in parallel, but that would increase power
dissipation and would reduce the LEDs brightness when turned on.

Thanks for any insight!

Koen

On Thu, 03 Jan 2008 13:45:07 +0100, Koen Vermeer <nospam@nospam.nl>
wrote:

>Hi,
>
>I have some basic knowledge of electronics, but I'm having trouble trying
>to get some real design to work. One of the things that do not work as
>expected is the BS170 MOSFET I'm trying to use. I know, it's not a very
>sophisticated part, but I have a few of them lying around and I thought
>they would be able to assist my learning.
>
>Anyway, I connect the gate and source of the BS170 both to the ground,
>and the drain to a 330 ohm resistor, connected to 3.6 V. When I then
>measure the potential of the drain, it's at 2.85 V, which means it draws
>a current of (3.6-2.85)/330 = 2.3 mA. Thus, the drain-source resistance
>is about 1.2 kohm. My expectation was a much larger 'off' resistance (say
>100 kohm or more).

The drain current should be tiny in this setup. Idss is spec'd at 0.5
uA max and should actually be a lot less, nanoamps maybe; the fet
should be off, not conducting at all. So something is wrong. Maybe
it's connected wrong, maybe it's zapped,

>
>I cannot relate anything on the data sheet (http://www.fairchildsemi.com/
>ds/BS%2FBS170.pdf) to my observation.

Idss is the relevant parameter.

John

On Thu, 03 Jan 2008 07:19:47 -0800, John Larkin wrote:

> The drain current should be tiny in this setup. Idss is spec'd at 0.5 uA
> max and should actually be a lot less, nanoamps maybe; the fet should be
> off, not conducting at all. So something is wrong. Maybe it's connected
> wrong, maybe it's zapped,

My thoughts were that the current should be very small as well, but it
isn't. It's quite stupid. I just tried two others: One was about the
same, the other gives a current of about 350 uA. Still way too large. It
doesn't make sense to me.

I can try changing the connections, but I guess that's a sure way to
break them...

Koen

On Thu, 03 Jan 2008 18:03:24 +0100, Koen Vermeer <nospam@nospam.nl>
wrote:

>On Thu, 03 Jan 2008 07:19:47 -0800, John Larkin wrote:
>
>> The drain current should be tiny in this setup. Idss is spec'd at 0.5 uA
>> max and should actually be a lot less, nanoamps maybe; the fet should be
>> off, not conducting at all. So something is wrong. Maybe it's connected
>> wrong, maybe it's zapped,
>
>My thoughts were that the current should be very small as well, but it
>isn't. It's quite stupid. I just tried two others: One was about the
>same, the other gives a current of about 350 uA. Still way too large. It
>doesn't make sense to me.
>
>I can try changing the connections, but I guess that's a sure way to
>break them...
>
>Koen


Sounds like, from the original numbers, that you had the gate
connected to the drain.

John

On Thu, 03 Jan 2008 10:06:32 -0800, John Larkin wrote:

>>My thoughts were that the current should be very small as well, but it
>>isn't. It's quite stupid. I just tried two others: One was about the
>>same, the other gives a current of about 350 uA. Still way too large. It
>>doesn't make sense to me.
>>I can try changing the connections, but I guess that's a sure way to
>>break them...
> Sounds like, from the original numbers, that you had the gate connected
> to the drain.

I was thinking about that, but while the data sheet isn't very clear
about the gate and source, the drain is rather obvious. Of course, I can
try to change the connection, and hope that I didn't kill the BS170
itself earlier...

Koen

On Fri, 04 Jan 2008 10:20:06 +0100, Koen Vermeer wrote:

> I was thinking about that, but while the data sheet isn't very clear
> about the gate and source, the drain is rather obvious. Of course, I can
> try to change the connection, and hope that I didn't kill the BS170
> itself earlier...

OK, I bought some new BS170s, and the first one I tried worked as
expected. So I guess I had a bad batch or something like that. Anyway,
I'm happy to have regained my sanity

Thanks for the help!

Koen