Simplest latch imaginable

In article <408A360D.9090508@nospam.com>,
Fred Bloggs <nospam@nospam.com> wrote:

Why build an SCR when they can be bought in a TO-92 package?

+----o~~~~o------+------/\/\--------+----+
| | R1 | |
| | | /
| | | 10K
| | | /
| / e \
| 100K \| |
| / 2n2907 |--+
--- \ /| |
- 9V | c |
| | | c
| | / 470 | |/
| +----o o---/\/\---+--| 2n2222
| | sw | |\
| | / e
| === 10K |
| |0.47u / |
| |metal film \ |
| | | |
+----------------+------------------+----+
Click to expand...

--
--
kensmith@rahul.net forging knowledge
 
John Larkin wrote...
Winfield Hill wrote:

Ken Smith wrote...

Fred Bloggs <nospam@nospam.com> wrote:

+----o~~~~o------+------/\/\--------+----+
| | R1 | |
| | | /
| | | 10K
| | | /
| / e \
| 100K \| |
| / 2n2907 |--+
--- \ /| |
- 9V | c |
| | | c
| | / 470 | |/
| +----o o---/\/\---+--| 2n2222
| | sw | |\
| | / e
| === 10K |
| |0.47u / |
| |metal film \ |
| | | |
+----------------+------------------+----+


Why build an SCR when they can be bought in a TO-92 package?

Indeed, e.g., 2n5062, 16 cents each. X0202MA, 52 cents each.

Depending on the value of R1 above, the current through the
made-up-SCR may be rather high, and much of it goes through
the two transistor's bases. While a BJT base can take higher
current than one might imagine, this is not a good situation.

The '5062 might pop a wire bond firing a squib, too, but probably
not if powered by a regular 9-volt battery. But it's a sensitive-
gate thingie, and a little static could fire it.
Click to expand...
As in Fred's design, but this is easily handled with a parallel
resistor, and I'd also add an RC on the input. BTW, some TO-92
parts can handle more than you'd imagine. The '5062 is rated at
10A for one-half ac cycle. The X0202MA is beefier, rated at 25A
for 8.3ms (the suggested 2n2907 and 2n2222 could not manage this).

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
Winfield Hill wrote:
John Larkin wrote...

Winfield Hill wrote:


Ken Smith wrote...

Fred Bloggs <nospam@nospam.com> wrote:

+----o~~~~o------+------/\/\--------+----+
| | R1 | |
| | | /
| | | 10K
| | | /
| / e \
| 100K \| |
| / 2n2907 |--+
--- \ /| |
- 9V | c |
| | | c
| | / 470 | |/
| +----o o---/\/\---+--| 2n2222
| | sw | |\
| | / e
| === 10K |
| |0.47u / |
| |metal film \ |
| | | |
+----------------+------------------+----+


Why build an SCR when they can be bought in a TO-92 package?

Indeed, e.g., 2n5062, 16 cents each. X0202MA, 52 cents each.

Depending on the value of R1 above, the current through the
made-up-SCR may be rather high, and much of it goes through
the two transistor's bases. While a BJT base can take higher
current than one might imagine, this is not a good situation.

The '5062 might pop a wire bond firing a squib, too, but probably
not if powered by a regular 9-volt battery. But it's a sensitive-
gate thingie, and a little static could fire it.


As in Fred's design, but this is easily handled with a parallel
resistor, and I'd also add an RC on the input. BTW, some TO-92
parts can handle more than you'd imagine. The '5062 is rated at
10A for one-half ac cycle. The X0202MA is beefier, rated at 25A
for 8.3ms (the suggested 2n2907 and 2n2222 could not manage this).

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
Click to expand...
The miniature hobby ignitors fire at something like 100-200mA and 3V, so
the current through the readily available transistors will not be quite
high- the OP asked for "simplest" so the static protection is reduced to
a shunt across the ignitor until it is armed. I since moved the 100K in
series with the battery- to kill battery turn-on dV/dt- and placed the
cap in parallel with the R1+SCR- so that during the arm sequence, the
ignitor shunt is not pulled when the circuit fails to standoff full
battery voltage. Radio Shack used to carry a little 1A SCR in TO-92, but
no more.
 
Fred Bloggs wrote...
Radio Shack used to carry a little 1A SCR in TO-92, but no more.
Click to expand...
It may sometimes be useful to restrict designs to parts that
are available at a local Radio Shack, but if so perhaps this
should be stated, since it leads to stinted awkward designs,
teaching less than straightforward ideal design principles.
Any hobbiest can order parts from DigiKey, Mouser or Jameco.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
On Sat, 24 Apr 2004 06:30:11 -0700, mike wrote:

Fred Bloggs wrote:

Fred Bloggs wrote:



MTB wrote:

Hi,

I want to ignite the second stage of a model rocket by using a mercury
switch to detect the deceleration of the first stage motor stopping.
My elementary knowledge of electronics seems to indicate some sort of
"latch" since the switching action of the mercury switch may only be
"on" for a few milliseconds. This latch would have to turn on as soon
as the mercury switch did and then remain on. The "load" in the
circuit would be the second stage igniter (essentially a fuse) which,
when it blew, would turn the circuit off again.
Whole thing would be powered by a small 9V battery.

Since the last electronics I did was when discrete transistors were
the norm, my abilities are small. I seem to remember though that a
monostable might serve this purpose?

Question is, what is the simplest (and therefore the lightest) circuit
to achieve this?

Any experts out there care to help?

Rgds,

/\/\



Please view in a fixed-width font such as Courier.


7V
R1= ---------------
ignitor ignitor current

+----o~~~~o------+------/\/\--------+----+
| | R1 | |
| | | /
| | | 10K
| | | /
| / e \
| 100K \| |
| / 2n2907 |--+
--- \ /| |
- 9V | c |
| | | c
| | / 470 | |/
| +----o o---/\/\---+--| 2n2222
| | sw | |\
| | / e
| === 10K |
| |0.47u / |
| |metal film \ |
| | | |
+----------------+------------------+----+


I suppose it would help to make allowance for arming the thing- this
uses a two pole shunt- move to ARM position after battery voltage is
thrown ON:
Please view in a fixed-width font such as Courier.

7V
SHUNT R1= ---------------
ignitor current
+--o ARM o SAFE o---+
| | |
+-+--/\/\---+---o~~~~o--+---------/\/\--------+----+
| 100K ignitor | R1 | |
| | | /
| | | 10K
| | | /
| | e \
| | \| |
| | 2n2907 |--+
--- | /| |
- 9V | c |
| | | c
| | / 470 | |/
| +------o o----/\/\---+--| 2n2222
| | sw | |\
| | / e
| === 10K |
| |0.47u / |
| |metal film \ |
| | | |
+-----------------------+---------------------+----+


This is a disaster waiting to happen.
Add a fail-safe to make absotively, posilutely CERTAIN that the
second stage cannot ignite before the first stage.
Pressure switch, conductor that the first stage burns thru...
anyhting to keep the second stage from firing when you
bump it.

SCR does the same thing, saves the weight of a transistor.
Don't know how much energy the ignitor takes, but charging
that cap might be dangerous.

I prefer switches that fail safe when a wire breaks loose.

This one requires that the ignitor shunt work.

Having said all that, aren't there commercial solutions for two-stage
rockets?
Click to expand...
Actually, there's a homebrew solution known as cellophane tape. You
tape the second stage engine to the first and the ejection gasses
set it off. The idea is that with a zero delay ejection charge, the
rocket has little time to decellerate (and change attitude -
possibly pointing it toward the ground) before the second stage
fires.

Dedicated newsgroups? Check there. Your MUCH better off
with recommendations from people who are currently doing successfully
the exact same thing. Solutions from a bunch of engineers with no clue
as to the gotchas in model rocketry may not be optimum...or safe.
Be careful that you quest for minimum weight doesn't lead you to
maximum blindness when it goes up in your face. Safety first!!

mike
Click to expand...

--
Best Regards,
Mike
 
Winfield Hill wrote:
Fred Bloggs wrote...

Radio Shack used to carry a little 1A SCR in TO-92, but no more.


It may sometimes be useful to restrict designs to parts that
are available at a local Radio Shack, but if so perhaps this
should be stated, since it leads to stinted awkward designs,
teaching less than straightforward ideal design principles.
Any hobbiest can order parts from DigiKey, Mouser or Jameco.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
Click to expand...
Okay- well your point is taken on illustration of best design
implementation, but the restriction to "retail" parts illustrates
another ever present design principle, and that is doing well enough
with available stock.
 
On Sun, 25 Apr 2004 15:33:51 GMT, Garrett Mace wrote:

I've experimented quite a lot with home-brew rockets (including fuel
rods based on cellulose, i.e. toilet paper :), and I can assure you that
the flight stability of a rocket improves with a higher center of gravity.
The lower the center of gravity, the bigger the stabilizer fins will have
to be to prevent your rocket from going anywhere but up.
It's actually quite easy to understand. Suppose the center of gravity is
found at one third of the height, measured from the bottom. This means
that two thirds of the body surface is located above the c.o.g., and one
third below it. This causes even the smallest air flow with a horizontal
component to push the rocket further off its course - unless this surface
ratio is compensated by the surface of the fins at the bottom. With the
c.o.g. in the top half of the rocket, the rocket will automatically point
its nose in the direction of the air flow. With a sufficiently high
c.o.g., one doesn't even need stabilizer fins (like with those firework
rockets, which only have a long, lightweight tailstick at the bottom).

But that only works up to a certain weight...and usually with the weight
distributed evenly. During the initial boost, there isn't enough airflow to
stabilize the rocket; you are depending on rotational inertia and the launch
rod to keep the rocket stable at the beginning. If you have a very heavy
object up front, the rotational inertia of the rocket is very low, while the
force due to acceleration is very high. The slightest disturbance will cause
the thrust end of the rocket to kick out.

Well, I'm probably thinking of the batteries as being a bit too heavy, a
significant portion of the rest of the rocket's mass. If he can get the
special low-current igniters, then it might have a chance of working with a
small battery.
Click to expand...
The COG should be at least one body tube dia ahead of the COP -
center of pressure - for stability.
--
Best Regards,
Mike
 
On Sun, 25 Apr 2004 00:20:15 +0100, MTB wrote:

On Sat, 24 Apr 2004 10:24:05 -0500, "Garrett Mace"
g.ryan@macetech.com> wrote:


"MTB" <mtbrecyclebin@hotmail.com> wrote in message
news:9da81cc6.0404240059.5940133@posting.google.com...
Hi,

I want to ignite the second stage of a model rocket by using a mercury
switch to detect the deceleration of the first stage motor stopping.
My elementary knowledge of electronics seems to indicate some sort of
"latch" since the switching action of the mercury switch may only be
"on" for a few milliseconds. This latch would have to turn on as soon
as the mercury switch did and then remain on. The "load" in the
circuit would be the second stage igniter (essentially a fuse) which,
when it blew, would turn the circuit off again.
Whole thing would be powered by a small 9V battery.

Since the last electronics I did was when discrete transistors were
the norm, my abilities are small. I seem to remember though that a
monostable might serve this purpose?

Question is, what is the simplest (and therefore the lightest) circuit
to achieve this?

Any experts out there care to help?

Rgds,

/\/\


Don't do this, for the following reasons:

1. Mercury switches will probably slosh up and trigger right when the rocket
is initally fired
2. Mechanical switches might do the same
3. Neither switch will sense the deceleration when the rocket motor burns
out: rockets coast for quite a while
4. If the switch is sensitive enough, then it will trigger on the launch pad
if you bump or tilt the rocket the wrong way: very unsafe!

Multistage model rockets use a primary engine designed for that purpose,
without an ejection charge in the top. The fuel burns to the top, and a jet
of flame goes up and ignites the second rocket. This accomplishes the task
very simply and well.

This link shows how to ignite the second engine if it is far away from the
first engine; up to about ten inches:
http://www.apogeerockets.com/education/how_to_multi-stage.asp In addition to
running a narrow tube from the first to the second rocket, you need to
provide vents at the top so that the hot exhaust reaches the second engine.


Garrett,

The narrow tube method is notoriously unreliable, hence the search for
something that is "guaranteed" to set off the second stage motor.
Trouble is, its almost never possible to guarantee it! The idea of a
timer seems OK to me, but doesn't allow for ease of alteration when a
different first stage motor is experimented with. Deceleration should
occur in any configuration!

Rgds,

/\/\
Click to expand...
Then if the 2nd engine is too far from the 1st to ensure reliable
ignition, the SCR circuit could be modified so that a fuse wire
pulls the gate low and when the zero delay ejection charge fires it
melts the fuse. The gate would then be pulled high through a
resistor to V+.


--
Best Regards,
Mike
 
On Mon, 26 Apr 2004 14:03:19 GMT, Fred Bloggs wrote:

Winfield Hill wrote:
Fred Bloggs wrote...

Radio Shack used to carry a little 1A SCR in TO-92, but no more.

It may sometimes be useful to restrict designs to parts that
are available at a local Radio Shack, but if so perhaps this
should be stated, since it leads to stinted awkward designs,
teaching less than straightforward ideal design principles.
Any hobbiest can order parts from DigiKey, Mouser or Jameco.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)


Okay- well your point is taken on illustration of best design
implementation, but the restriction to "retail" parts illustrates
another ever present design principle, and that is doing well enough
with available stock.
Click to expand...
I'd be hard pressed to find an SCR at any of the 3 Radio Sheiss
stores around here. There's no double sided PCB, no 741s, 555s, ...
just some dolt that says, "Can I help you."

--
Best Regards,
Mike
 
Fred Bloggs wrote...
Okay- well your point is taken on illustration of best design
implementation, but the restriction to "retail" parts illustrates
another ever present design principle, and that is doing well enough
with available stock.
Click to expand...
Yes, I'm simply suggesting that you specify when that's the goal.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
I reckon Fred's got it right - in the given application it does actually
work. I have put many discrete SCRs into products, they are extremely cheap.
4c or so buys a dual npn/pnp transistor, but not an scr.....
 
Terry Given wrote...
I reckon Fred's got it right - in the given application it does actually
work. I have put many discrete SCRs into products, they are extremely cheap.
4c or so buys a dual npn/pnp transistor, but not an scr.....
Click to expand...
You may feel good about it but I don't. Fred talks about 200mA currents,
and yet the ABSOLUTE maximum spec for a 2N2222 base current is 200mA.
That's too close for me, I prefer the SCR's 10A max spec safety margin.

And surely 16 cents cannot be an issue. BTW, that's 16 cents in small
quantity from a major distributor, let's talk apples and apples on this.
The price for plastic versions of Fred's two parts comes to 2x 9 cents
= 18 cents total, qty 10, from the same place.

And what about the cost of all those resistors and the assembly costs?
What about space and weight in the rocket?

Sorry, the 2n5061, 2n5064, etc., SCR appears to win here on all counts.

Now, as for using a dual npn/pnp transistor four-layer "SCR" at low
currents, yes of course. They make great flip flops, and I like the
way you can reach inside with another transistor and set or clear it.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
Winfield Hill wrote...
And surely 16 cents cannot be an issue.
Click to expand...
Aha! I found them at 7.8 cents, so there!

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
On Wed, 28 Apr 2004 09:11:35 GMT, the renowned Fred Bloggs
<nospam@nospam.com> wrote:

You may feel good about it but I don't. Fred talks about 200mA currents,
and yet the ABSOLUTE maximum spec for a 2N2222 base current is 200mA.
That's too close for me, I prefer the SCR's 10A max spec safety margin.

There is no such rating- even the P2N2222A from ON is rated at 600mA IC
continuous- and Fairchild rates theirs at 1.0A.
Click to expand...
Here is one with such a rating:
http://www.semiconductors.philips.com/acrobat/datasheets/2N2222_CNV_2.pdf

Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
 
Spehro Pefhany wrote...
Fred Bloggs wrote:

There is no such rating- even the P2N2222A from ON is rated at 600mA IC
continuous- and Fairchild rates theirs at 1.0A.

Here is one with such a rating:
http://www.semiconductors.philips.com/acrobat/datasheets/2N2222_CNV_2.pdf
Click to expand...
Thanks, Spef, I was looking at the detailed Philips datasheet.
Manufacturers often leave out the Ib max spec, but one should
not simply substitute the higher Ic max spec in its place.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
Spehro Pefhany wrote:
On Wed, 28 Apr 2004 09:11:35 GMT, the renowned Fred Bloggs
nospam@nospam.com> wrote:



You may feel good about it but I don't. Fred talks about 200mA currents,
and yet the ABSOLUTE maximum spec for a 2N2222 base current is 200mA.
That's too close for me, I prefer the SCR's 10A max spec safety margin.

There is no such rating- even the P2N2222A from ON is rated at 600mA IC
continuous- and Fairchild rates theirs at 1.0A.


Here is one with such a rating:
http://www.semiconductors.philips.com/acrobat/datasheets/2N2222_CNV_2.pdf

Best regards,
Spehro Pefhany
Click to expand...
Please view in a fixed-width font such as
Courier.



CRY-BABY MOLLIFIER



---+--------+
| |
| \
| 10k
| \
e /
\| |
2N2907A |------+
/| |
c |
| 1N5817 |
| - |
+--|>|---+
| - |
| |
| c
| |/
+------| 2N2222A
| |\
\ e
10k |
\ |
/ |
| |
----+--------+
 
"Fred Bloggs" <nospam@nospam.com> schreef in bericht
news:408F7535.8030700@nospam.com...
You may feel good about it but I don't. Fred talks about 200mA
currents,
and yet the ABSOLUTE maximum spec for a 2N2222 base current is 200mA.
That's too close for me, I prefer the SCR's 10A max spec safety margin.

There is no such rating- even the P2N2222A from ON is rated at 600mA IC
continuous- and Fairchild rates theirs at 1.0A.
Click to expand...
Base current?


--
Thanks, Frank.
(remove 'x' and 'invalid' when replying by email)
 
Winfield Hill wrote:
Spehro Pefhany wrote...

Fred Bloggs wrote:


There is no such rating- even the P2N2222A from ON is rated at 600mA IC
continuous- and Fairchild rates theirs at 1.0A.

Here is one with such a rating:
http://www.semiconductors.philips.com/acrobat/datasheets/2N2222_CNV_2.pdf


Thanks, Spef, I was looking at the detailed Philips datasheet.
Manufacturers often leave out the Ib max spec, but one should
not simply substitute the higher Ic max spec in its place.
Click to expand...
That Ib,max derives from base junction power dissipation and not the
current handling capability- does current somehow jump from collector to
emitter without going through the base?
 
Frank Bemelman wrote:
"Fred Bloggs" <nospam@nospam.com> schreef in bericht
news:408F7535.8030700@nospam.com...


You may feel good about it but I don't. Fred talks about 200mA

currents,

and yet the ABSOLUTE maximum spec for a 2N2222 base current is 200mA.
That's too close for me, I prefer the SCR's 10A max spec safety margin.

There is no such rating- even the P2N2222A from ON is rated at 600mA IC
continuous- and Fairchild rates theirs at 1.0A.


Base current?
Click to expand...
For what it's worth- calculations show 100mA max in this application.
 
Fred Bloggs wrote...
CRY-BABY MOLLIFIER

---+--------+
| |
| \
| 10k
| \
e /
\| |
2N2907A |------+
/| |
c |
| 1N5817 |
+--|>|---+
| |
| c
| |/
+------| 2N2222A
| |\
\ e
10k |
\ |
/ |
| |
----+--------+
Click to expand...
The lengths one will go to to avoid using the proper part.

SHUNT
+--o ARM o SAFE o---+ [ mollifier added ]
| | |
+-+--/\/\---+---o~~~~o--+---------/\/\--------+------+
| 100K | ignitor R1 | |
| | | 10K
| | e |
| | \| |
| | 2n2907 |----+
--- | /| |
- 9V | c |
| | 1n5817 | |
| | +-|>|--+
| | | |
| | / 470 | |/
| +------------------o o----/\/\---+----| 2n2222
| | sw | |\
| | 0.47u | e
| === 10K |
| | metal film | |
+-----------+---------------------------------+------+
Click to expand...

vs. this smaller, cheaper, better performing, more reliable,
more simple, more sensible SCR version of Fred's circuit:

.. SHUNT
.. +--o ARM o SAFE o---+
.. | | |
.. +-+--/\/\---+---o~~~~o--+------/\/\--------,
.. | 100K | ignitor R1 |
.. | | _|_
.. --- | _\_/_
.. - 9V | / 220 /|
.. | +----------o o----/\/\---+--' | 2n5061
.. | | sw | | 2n5064
.. | === 0.47u 470 |
.. | | metal film | |
.. +-----------+-------------------------+----'

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top