This item is for the discussion of transistors and related components.

16 responses total.

I recently decided to try to use a MOSFET for an application, but have had
a lot of trouble finding information for my current level of
understanding. I asked first in diy cf, but didn't find many knowledgeable
participants. Here is how the discussion went so far in the Electronics
item in diy:

  #175 of 179: by Rane Curl (rcurl) on Wed, Mar 28, 2001 (00:00):  I am
 installing a sump pump in my basement, using a "bilge pump", as when
 backwater comes up the floor drain it will be very shallow as it starts
 to spread, and a bilge pump will work with less than an inch of depth. I
 thought of using a microswitch and float as a sensor, but have decided
 that a conductivity switch would be more reliable.  The pump will be
 switched on with a 30 A relay using 12 V. It draws ca. 150 ma. I would
 like to switch that with the simplest electronic system possible, and 
 that appears to be a single MOSFET. 

 For that purpose I need an "enhacement mode" (normally off) MOSFET, so as
 not to have an additional G bias circuit. My problem is that electronic
 parts catalogs that I have (Mouser, DigiKey, radioshack) do not indicate
 whether MOSFETS are "normally off". I plan on using a P-channel MOSFET
 like the VP0300L, but don't know if it is "normally off".

 The circuit will have the MOSFET S to +12 V, D to the relay and then to
 (-)  ground, and G to to both +12V with a large resistor and to the
 sensor probe to contact water (which will be at (-) ground).

 What MOSFETS with other characteristics like the VP0300L are "normally
 off"?

  #176 of 179: by Slower traffic keep right (gull) on Wed, Mar 28, 2001
 (13:06):
 Re #175: Silly question:  If the plan is to pump water out if it backs
 up from the sewer, where are you going to pump it?  The sewer's already
 full.

  #177 of 179: by Rane Curl (rcurl) on Wed, Mar 28, 2001 (14:09):
 Outside.

  #178 of 179: by klaus (n8nxf) on Wed, Mar 28, 2001 (14:27):
 I should know more about MOSFETS than I do.  I'm pretty sure that most of
 them are normally off.  I noticed that Digi-Key has data sheets on their
 web site that you can access once you find a part that you think might
 work.  That would probably be helpful in making a selection.

  #179 of 179: by Rane Curl (rcurl) on Wed, Mar 28, 2001 (14:43):
 The 2001 ARRL Handbook has two tables of JFETS and MOSFETS, for signal
 and power FETS. However, most of the devices they list are not sold by
 DigiKey, Mouser or RadioShack, and there are errors in the table 
 headings. One table, though, does show the gate cutoff voltage for
 N-Channel signal FETS, with about an equal number of normally on and
 normally off devices. Thanks for the DigiKey suggestion - I was going to
 turn to the web next, after I had "milked" the local experts for info.
 8^}

So, that is the situation so far, but I have learned lots more, and now
understand the operation of a wide variety of devices, such as listed
in the title of this item.

What I have purchased is a VMOS MOSFET, P-channel, enhancement mode, rated
for 1 A Id, 100 Vds max - from an electronic surplus company, and will try
that. Their catalog is the first one that mentioned "enhancement mode" (as
opposed to "depletion mode"), and I learned about P- and N- channel
MOSFETs from several sources, but VMOS was a stumper. I found an
explanation of VMOS via the WWW, which simply put is a "V" configuration
for the conduction layer of a MOSFET, to provide greater current capacity. 

I have to admit that one attraction of MOSFETS to me is that there
characteristics are rather like triode tubes, which I understand!  In
particular they have very low gate (grid) current (high input impedance,
and mostly capacitive) and moderate gain. 

This search, besides finding a component I think I can use, gave me a
narrow look into the world of solid-state devices: what a variety there is
of them, how ingenious they are, and how rapidly the devices are still
evolving, and how antiquated are many sections of the ARRL Handbook (which
describes MOSFETs, but not VMOS, DMOS or HEXFET, even though a DMOS is
used in one of the circuits that is described)! 

(CMOS is something rather different.....)

I found that cool about FETs, too...though my reactions were reversed -- 
started learning about tubes, and realized, 'gee, these are just like 
MOSFETs!' ;>

(I learned tubes when the only solid-state diodes were selenium (not
long after the only ones were pyrite crystals....)).

My MOSFET breadboard setup to control a relay from a high impedance
source - works. This uses the same P-channel enhancement mode VMOS I
cited above. The high impedance of the input is nice as I can put
in delay for the on/off response with just a capacitor. (P.S. It
isn't really a break board - it a Radio Shack plastic "bread board"
with lots of tiny holes connecting lots of circuit traces for lots
of tiny part. I actually used a *bread board* in my early days of
electronics. The term still sticks: "breadboarding" a circuit. Now it is
more like surfboarding a circuit.)

Is there a reason you didn't use an SCR or Triac for this?

I'm controlling DC, for one thing. Second, the MOSFET draws no
gate current (in DC use), and my source resistance is very high. 

Another feature is that no reverse bias is required for the
enhancement-mod MOSFET - that is, it is "normally off".

That's a good thing.  I had been thinking about how you would
work having a bias voltage on there, to assure that it was off.
(Sounds like they've made it internal.)

It is inherent. There is no conduction path between the source and
drain until the gate induces (not injects) some charge carries into the
region. 

Here is the problem. A device produces a +2.5 volt signal w.r.t.  a ground
(2 ma max). I would like that to turn on a 9-12 volt supply (w.r.t the
same ground) with a circuit that draws almost no current in standby. I've
been trying to imagine coupled MOSFETS, but everything depends upon
threshold voltages - the "curves" for the devices, which I don't have.
Can someone help me think this through?

Further to the above, what is the best "transistor manual" that gives
characteristic curves for low power transistors, including MOSFETS? I
think I have drafted a circuit that would use a grounded emitter
transistor or N-channel MOSFET to drive a P-channel MOSFET controlling the
+V power source. However specific characteristics are needed to choose
devices and circuit components for low standby current and low MOSFET
threshold voltage.

[i.e., what is the transistor equivalent to the RCA Receiving Tube Manual?
(That should date me....)]

Various semiconductor manufacturers release manuals for their parts, 
though I don't know if any are available on the web.  One that I've 
used before in print form is the ECG Semiconductor Replacement Manual, 
but that's more of a cross-reference.  (Though it does have some parts 
specs.)

I'm still trying to figure out what a w.r.t / w.r.t. is.

With respect to what?

ECG is now NTE, but hopefully they haven't changed their manuals that
much.

As for finding things about the characteristics of semis, I'd say look
at MCM Electronics, they have a couple books available.

I've asked my question on an electronics List, and gotten good advice. 
I'm going to go with a comparator IC that has very low leakage currents
and allows voltage level shift: input can be 0-3 volts, and control up to
15 volts. This should have no trouble gating a MOSFET. The nice thing
about this choice is that both the driver and the MOSFET are essentially
on-off, so I don't need to get any transisitor characteristics! Still, a
lot of responders suggested the book _Ther Art of Electronics_, by
Horowitz and Hill, so I'm buying a second-hand copy. 

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss