Maker Pro
Maker Pro

Very low thresold/depletion power TO-220 MOSFET?

Hi all,

The only missing link in a design I'm doing is a swith that must be
able to operate at very low voltages (1V or less if possible), have
minimal losses, available through mouser/digikey/etc, and be
through-hole (TO-220, say) for breadboard prototyping. I've been out of
the field for some time so I don't know what's available. Here are the
specs:

rON max @ 500mA : 100mOhms
Vgs(th) : as small as possible, less than 1V
Vgs(ON): less than 1V
Imax, power, etc : should be able to handle 500mA when ON.

frequency is irrelevant. The switch is going to be turned on or off
every few seconds at the fastest. Relays don't cut it because the
circuit is going to be constantly shaking!

Bipolar darlingtons (I would need them because I also don't have much
power available to turn the switch ON/OFF) seem to be out of the
question since their Vce(sat) is 1V or more, translating to an rON of
about 2 ohms. Their Vbe(on) is also not that great.

I found this Vishay P-channel MOSFET (SUP/SUB15P01-52) which comes
close, and that I *could* use at the cost of 2 more transistors and
resistors in the design. Unfortunately there seems to be no N-channel
equivalents. From what I see there are no IGBTs that cut it due to
Vce(sat).

Strictly speaking the switch should be 'on' by default, but I couldn't
find any depletion mode MOSFETs that would satisfy the Ron and
through-hole specs, that's why I am look for very low threshold
enhancement mode transistors.

Are the specs hopeless, or does anybody know of something useable?

Best,

- Cesar
 
P

Phil Allison

Jan 1, 1970
0
The only missing link in a design I'm doing is a swith that must be
able to operate at very low voltages (1V or less if possible), have
minimal losses, available through mouser/digikey/etc, and be
through-hole (TO-220, say) for breadboard prototyping. I've been out of
the field for some time so I don't know what's available. Here are the
specs:

rON max @ 500mA : 100mOhms
Vgs(th) : as small as possible, less than 1V
Vgs(ON): less than 1V
Imax, power, etc : should be able to handle 500mA when ON.


** The ZTX1049A meets your spec.

http://www.farnell.com/datasheets/38789.pdf

Vce sat = 30mV @ 0.5 amp (= 60 mOhms)

This with Ib a mere 10 mA and Vbe about 0.8 volts.

Tiny TO92 pack.

Cheap.



......... Phil
 
K

Ken Smith

Jan 1, 1970
0
Hi all,

The only missing link in a design I'm doing is a swith that must be
able to operate at very low voltages (1V or less if possible), have
minimal losses, available through mouser/digikey/etc, and be
through-hole (TO-220, say) for breadboard prototyping. I've been out of
the field for some time so I don't know what's available. Here are the
specs:

rON max @ 500mA : 100mOhms
Vgs(th) : as small as possible, less than 1V
Vgs(ON): less than 1V
Imax, power, etc : should be able to handle 500mA when ON.

I'd look at the Supertex web site. They make depletion mode N channel
MOSFETs. Can you make something like that work?
 
That's a very interesting transistor, thanks! The gain is almost where
I want it but it should be doable. Printing datasheet now...

- Cesar
 
P

Phil Allison

Jan 1, 1970
0
That's a very interesting transistor, thanks! The gain is almost where
I want it but it should be doable. Printing datasheet now...


** Please ALTER your settings on Google Groups so that QUOTED text becomes
visible.

It is VERY bad etiquette to post on usenet with no visible hint as to whom
or which words to you are addressing.



........ Phil
 
C

Cesar Crusius

Jan 1, 1970
0
Phil said:
** Please ALTER your settings on Google Groups so that QUOTED text becomes
visible.

It is VERY bad etiquette to post on usenet with no visible hint as to whom
or which words to you are addressing.

....... Phil

Thanks. I've been around the net for some time and know it, so I ALSO
AVOID SCREAMING. I always make sure the default behavior on my
mail/news/etc applications is to quote the post.

To tell you the truth I also found the google groups behavior very
annoying but saw no way of changing it. If you can point me to where in
the settings I can change to quoted default I'd do it immediately, but
there doesn't seem to be a way.

Being annoyed myself, though, I searched some and now I know how to do
it: It turns out that on Google groups the only way to quote the
article is *not* to click on "reply", but instead to click on "show
options" and then on the "reply" that shows after that. Go figure. At
least I know how to do it now though.

- Cesar
 
P

Phil Allison

Jan 1, 1970
0
"Cesar Crusius"
Thanks. I've been around the net for some time and know it, so I ALSO
AVOID SCREAMING. I always make sure the default behavior on my
mail/news/etc applications is to quote the post.

To tell you the truth I also found the google groups behavior very
annoying but saw no way of changing it. If you can point me to where in
the settings I can change to quoted default I'd do it immediately, but
there doesn't seem to be a way.

Being annoyed myself, though, I searched some and now I know how to do
it: It turns out that on Google groups the only way to quote the
article is *not* to click on "reply", but instead to click on "show
options" and then on the "reply" that shows after that. Go figure. At
least I know how to do it now though.

** Good one.

Now I can give explicate advice to others in the same boat.

( SS Google ? )


........... Phil
 
W

Winfield Hill

Jan 1, 1970
0
Cesar Crusius wrote...
I searched some and now I know how to do it: It turns out that
on Google groups the only way to quote the article is *not* to
click on "reply", but instead to click on "show options" and
then on the "reply" that shows after that.

Thanks for that useful factoid. Sometimes I can't get my usenet
provider's system to cooperate, or it has lost an article, etc.,
and I have to use Google -- so that's a great usage hint you've
figured out and given us!
 
C

Cesar Crusius

Jan 1, 1970
0
Phil said:
** The ZTX1049A meets your spec.

http://www.farnell.com/datasheets/38789.pdf

Vce sat = 30mV @ 0.5 amp (= 60 mOhms)

This with Ib a mere 10 mA and Vbe about 0.8 volts.

Tiny TO92 pack.

Cheap.

I just bumped into some solid state relays that may just be the ticket.
The PS710B-1A from NEC has almost the specs I want. NEC has some "on by
default" relays but not with low resistance. I shall search some
more...

- Cesar
 
W

Winfield Hill

Jan 1, 1970
0
Cesar Crusius wrote...
I just bumped into some solid state relays that may just be the ticket.
The PS710B-1A from NEC has almost the specs I want. NEC has some "on
by default" relays but not with low resistance. I shall search some
more...

Those are made with 60V depletion-mode FETs. Sadly, there's not a
lot of choice with such beasts. When I found the PS710 and saved its
datasheet, a few months ago, I saved it as "ps710A_0.1-ohm-SSR.pdf"
adding the low-Ron description to the part number, which I rarely do.
 
C

Cesar Crusius

Jan 1, 1970
0
Winfield said:
Those are made with 60V depletion-mode FETs. Sadly, there's not a
lot of choice with such beasts. When I found the PS710 and saved its
datasheet, a few months ago, I saved it as "ps710A_0.1-ohm-SSR.pdf"
adding the low-Ron description to the part number, which I rarely do.

Yep. Sadly, there seems to be nothing I can use. I was reviewing the
circuit requirements and it turns out nothing that is not normally
closed would be ideal, as now it seems I actually need the switch
closed at about 0.2V. Depletion mode FETs are rare enough, and when you
get them Ron is too high. I kind of threw in the towel and started
searching for normal relays - those one can easily get normally closed,
but alas, they require way too much current to turn on (off). The
search continues, but I may need to bump up that voltage somehow...

- Cesar
 
P

Phil Hobbs

Jan 1, 1970
0
Cesar said:
Winfield Hill wrote:




Yep. Sadly, there seems to be nothing I can use. I was reviewing the
circuit requirements and it turns out nothing that is not normally
closed would be ideal, as now it seems I actually need the switch
closed at about 0.2V. Depletion mode FETs are rare enough, and when you
get them Ron is too high. I kind of threw in the towel and started
searching for normal relays - those one can easily get normally closed,
but alas, they require way too much current to turn on (off). The
search continues, but I may need to bump up that voltage somehow...

- Cesar

Since you're not switching that often, perhaps you could use a small
boost regulator to generate the gate bias you need.

Cheers,

Phil Hobbs
 
W

Winfield Hill

Jan 1, 1970
0
Cesar Crusius wrote...
Yep. Sadly, there seems to be nothing I can use. I was reviewing the
circuit requirements and it turns out nothing that is not normally
closed would be ideal, as now it seems I actually need the switch
closed at about 0.2V. Depletion mode FETs are rare enough, and when you
get them Ron is too high. I kind of threw in the towel and started
searching for normal relays - those one can easily get normally closed,
but alas, they require way too much current to turn on (off). The
search continues, but I may need to bump up that voltage somehow...

Maybe a latching relay? What are you working on?

Actually, before leaving depletion-mode FETs behind, did you
consider low-Ron JFETs. While meant for signal low-resistance
switching, they can work great in power applications. For
example, the J105 at 3 ohms max, perhaps 1.5 ohms typical, in
a tiny package. Maybe a bunch in parallel?
 
R

Rich Grise, but drunk

Jan 1, 1970
0
Cesar Crusius wrote...

Thanks for that useful factoid. Sometimes I can't get my usenet
provider's system to cooperate, or it has lost an article, etc., and I
have to use Google -- so that's a great usage hint you've figured out and
given us!

Don't forget "View as Tree". ;-)

Cheers!
Rich
 
K

Ken Smith

Jan 1, 1970
0
Winfield Hill said:
Actually, before leaving depletion-mode FETs behind, did you
consider low-Ron JFETs. While meant for signal low-resistance
switching, they can work great in power applications. For
example, the J105 at 3 ohms max, perhaps 1.5 ohms typical, in
a tiny package. Maybe a bunch in parallel?

Take a look at Interfet for JFETs with low RDS(on) values. They make some
parts that are interesting.
 
Top