Transistor switch design problem

[todd fahrner]

First: I'm a complete electronics novice, in over my head. Apologies
in advance for poor descriptions and follow-up questions.

The overall problem: I have a Hall effect sensor (in a throttle).
There's a main, a ground, and a feedback wire. In the rest state, the
feedback wire returns ~5V. As the throttle is opened, the feedback
wire voltage drops progressively. I wish to close a switch powering a
~9V component when the throttle feedback voltage drops below 4V, and
open it as the voltage rises again above it.

All of the discussions and diagrams of transistor switches I've found
work by saturating a transistor to close a switch; i.e.; rising V
would do it. I need to have falling V do it. Seems like a simple
inversion should be possible, but I can't figure out how. Hints?
Alternative suggestions? Thanks.

 

[Homer.Simpson]

todd fahrner said

First: I'm a complete electronics novice, in over my head. Apologies
in advance for poor descriptions and follow-up questions.

The overall problem: I have a Hall effect sensor (in a throttle).
There's a main, a ground, and a feedback wire. In the rest state, the
feedback wire returns ~5V. As the throttle is opened, the feedback
wire voltage drops progressively. I wish to close a switch powering a
~9V component when the throttle feedback voltage drops below 4V, and
open it as the voltage rises again above it.

All of the discussions and diagrams of transistor switches I've found
work by saturating a transistor to close a switch; i.e.; rising V
would do it. I need to have falling V do it. Seems like a simple
inversion should be possible, but I can't figure out how. Hints?
Alternative suggestions? Thanks.

Try a PNP transistor:


5V 9V
/ |
|< |
Sig---R--< PNP LOAD
|- |
\ |
| |
R /
| |-
---R--< NPN
| |<
R \
| |
| |
gnd gnd


Hope the load isn't a Nitrous bottle. ;-)
This is very crude.

Sorry for the poor ASCII. This is my first!


--
I drive way too fast to worry about cholesterol.

 

[John Fields]

On 18 Jul 2004 10:36:40 -0700, fahrner@pobox.com (todd fahrner) wrote:

First: I'm a complete electronics novice, in over my head. Apologies
in advance for poor descriptions and follow-up questions.

The overall problem: I have a Hall effect sensor (in a throttle).
There's a main, a ground, and a feedback wire. In the rest state, the
feedback wire returns ~5V. As the throttle is opened, the feedback
wire voltage drops progressively. I wish to close a switch powering a
~9V component when the throttle feedback voltage drops below 4V, and
open it as the voltage rises again above it.

All of the discussions and diagrams of transistor switches I've found
work by saturating a transistor to close a switch; i.e.; rising V
would do it. I need to have falling V do it. Seems like a simple
inversion should be possible, but I can't figure out how. Hints?
Alternative suggestions? Thanks.

+12>-----+----------------+-------+------------+----------+
| | | | |
| | [10K] | |
[330R] | | | |
| +-----[1M]--+ | |
| | | | | |
VIN>--------[10K]-----+--|+\ | E |K
| | >-----+--[3K3]----B PNP [1N5359]
+------+ +--|-/ C |
| | |1/2 LM393 | |
|K | | +----------+ |
[Z5.1V] [10K]<--+ |K | |
| | |+ [1N4001] [RELAY COIL] |
| | [1ľF] | | |
| | | | | |
GND>-----+------+-----+-------------+----------+----------+


Adjust the pot for 4V at the - input of the comparator, then when VIN
falls below 4V the output of the comparator will go low, turning on
the transistor and the relay.

--
John Fields

 

[Robert C Monsen]

"todd fahrner" <fahrner@pobox.com> wrote in message
news:844274b5.0407180936.14e51820@posting.google.com...

First: I'm a complete electronics novice, in over my head. Apologies
in advance for poor descriptions and follow-up questions.

The overall problem: I have a Hall effect sensor (in a throttle).
There's a main, a ground, and a feedback wire. In the rest state,
the
feedback wire returns ~5V. As the throttle is opened, the feedback
wire voltage drops progressively. I wish to close a switch powering
a
~9V component when the throttle feedback voltage drops below 4V, and
open it as the voltage rises again above it.

All of the discussions and diagrams of transistor switches I've
found
work by saturating a transistor to close a switch; i.e.; rising V
would do it. I need to have falling V do it. Seems like a simple
inversion should be possible, but I can't figure out how. Hints?
Alternative suggestions? Thanks.

View in courier font (using notepad perhaps?)

5V --------------o-------------o---------.
| | |
.-. .-. |
| | | | 8.2k |
| | 100 | | |
'-' '-' |
| | |
,-o-. | |
4.7k | | | |
___ |< >| | |
Input -|___|--| |---------o |
|\ /| | |
| | | |
| | | |<
| o-------------------|
| | | |\
| | | |
| | | |
| | | o--------.
| | | | |
| .-. .-. | C| Relay
| | | 3.3k | | 33k - C|
| | | | | ^ C|
| '-' '-' | |
| | | | |
GND ------------o---o-----------o---------o--------o



Diode = 1N4148
PNPs = 2N3906
Relay = 5V switching

created by Andy´s ASCII-Circuit v1.24.140803 Beta www.tech-chat.de

This design assumes a regulated 5V supply.

When the input gets down to 4V, the differential amplifier will turn
on the rightmost transistor, causing the relay to be closed.

You could replace the differential amp with a comparator, but that
would be cheating...

Regards,
Bob Monsen

 

[John Popelish]

todd fahrner wrote:

First: I'm a complete electronics novice, in over my head. Apologies
in advance for poor descriptions and follow-up questions.

The overall problem: I have a Hall effect sensor (in a throttle).
There's a main, a ground, and a feedback wire. In the rest state, the
feedback wire returns ~5V. As the throttle is opened, the feedback
wire voltage drops progressively. I wish to close a switch powering a
~9V component when the throttle feedback voltage drops below 4V, and
open it as the voltage rises again above it.

All of the discussions and diagrams of transistor switches I've found
work by saturating a transistor to close a switch; i.e.; rising V
would do it. I need to have falling V do it. Seems like a simple
inversion should be possible, but I can't figure out how. Hints?
Alternative suggestions? Thanks.

The specific component designed to make such a one bit analog to
digital decision is called a comparator. It has two inputs and one
output. one input is labeled + and one is labeled -. When the +
input is more positive than the - input, the output goes positive,
which in most cases means that an output transistor connected to the
negative supply rail, turns off. The inputs can compare any voltages
that are in the common mode range, usually somewhere within the supply
rails. So, to use a comparator to make this decision, you need some
source that provides 4 volts to be compared to your hall effect
signal. The comparator will need a supply that surrounds the 4 volt
reference. Comparator outputs typically can carry between a couple
milliamps to up to about 30 milliamps. If you need more output
current than that, you will have to add a transistor to the output to
provide a higher current output.

Here is a data sheet for a common dual comparator and a single:
http://cache.national.com/ds/LM/LM193.pdf
http://cache.national.com/ds/LM/LM111.pdf

You can also use an opamp as a comparator in any application where the
output does not have to switch off or on in sub microsecond time.
Opamps are essentially comparators with push-pull output stages and
are slowed enough that the inputs cap hold the output in any
intermediate stage between full pull down and pull up.

Here is the data sheet for the opamp version based on the same front
end as the above dual comparator.
http://cache.national.com/ds/LM/LM158.pdf

These are very useful components to be familiar with.

--
John Popelish

 

[todd fahrner]

"Robert C Monsen" <rcsurname@comcast.net> wrote in message news:<HTAKc.89834$WX.27865@attbi_s51>...

This design assumes a regulated 5V supply.

Three replies, three approaches... Your design seems simplest, but
still a little confusing: my circuit design reading skills are below
fundamental. Concerning supply, I have been assuming that I could tap
the throttle feed wire without interfering with its function.

I'd say I'm experiencing a fight or flight reaction to the challenge
of learning this stuff well enough to solve my problem in a reasonable
timeframe.