security

mctrivia · 2009-02-16 20:12

is there such a device that: has 4 leads. pin a - b act as a wire but if power is applied between c and d then the link between a and b is permanently broken? want for a tamper proof system.

heater · 2009-02-16 21:04

Depends what you mean by "wire" and "permanently" broken.

Let's assume "a" is a source of power to something via "b". Use a simple fuse for the connection "a" to "b".

Now when some action happens at "c" or "d" we will blow the fuse permanently. Lets just use "c" and have it connected to the base of a transistor. The collector of which goes to "b" and the emitter to ground. When "c" goes high the transistor comes on and shorts the fuse to ground "poof" connection permanently broken. Use NPN transistor in this case, rated to blow the fuse first. Use a resistor in the base connection to limit current into the base. Or use a Triac. Or use a relay. This is called a "crowbar" for reasons that should be obvious.

Tell us more about what is connected up here.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

Beau Schwabe · 2009-02-16 21:42

mctrivia,

I'm not sure if a solid state device exists, but you could put something together pretty easily... I'm sure that·you can get a surface mount or small pig-tail fuse ... Pico-Fuse ... comes to mind.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

dMajo · 2009-02-16 22:00

Could you explain better your application purposes: maybe we have a better solution

Is this going on your ds32c35+prop board?

mctrivia · 2009-02-16 22:05

it is for the rx tx between to ic

mctrivia · 2009-02-16 22:08

I do have some 250ma fuses

dMajo · 2009-02-16 22:15

mctrivia said...
it is for the rx tx between to ic

do you want to make a one time programming feature (eg. burn the dtr lead at first run)?

heater · 2009-02-16 22:26

Nice picture Beau. Just what I had in mind. Of course, as you surely know, the transistor will always blow first to protect the fuse[noparse]:)[/noparse]

Now for signal carrying "wire" this gets a bit more delicate.

Well, you could just blow the Tx output or Rx input if your control voltage is high enough and you really mean permanent!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

mctrivia · 2009-02-16 23:08

I don't want it to be 1 time only but i do want to be able to encase the entire package in resin and be able to blow out the fuse when I have finished testing so the code can not be altered. I thought about just putting a fuse in series forcing the pin to low output and applying high voltage to the input but I don't want to risk damaging the prop as I only have 1 at the moment. Is there a safe voltage that would work but not damage prop?

mctrivia · 2009-02-16 23:11

looking at the image that would probably work. transistor would not effect data and could tight both a and c high to blow fuse.

Carl Hayes · 2009-02-17 00:22

Well, there are ratchetng relays. Energize the coil once, and the contacts open. Energize it again, and they close. The cycle repeats.

If you used such a relay, DPST, and used one set of contacts for your A-B and connected the coil through C-D and a pushbutton, then when the pushbutton was pushed the relay would switch A-B off, as you desire -- and also swould switch C-D off so that pushing the button again would do nothing.

To reset it you'd have to short C-D and push the button.

There are many other ways to do it, but few so simple and reliable.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net

Carl Hayes · 2009-02-17 00:24

Beau Schwabe (Parallax) said...
mctrivia,

I'm not sure if a solid state device exists, but you could put something together pretty easily... I'm sure that·you can get a surface mount or small pig-tail fuse ... Pico-Fuse ... comes to mind.

Much better to use an SCR instead of a transistor.· Such a circuit is called a crowbar and is a standard approach.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net

mctrivia · 2009-02-17 01:02

ya relay is not practical to big. will see if I have any scr's to test it out with. i know I have npn transisters.

Any reason scr is better then a transistor?

NPN FMMT495TA - 1A Ic
SCR P0102BL 5AA4 - 800mA

The SCR costs twice as much.

Beau Schwabe · 2009-02-17 01:35

Carl Hayes,

True, the SCR would ensure that the event had completed... [noparse]:o[/noparse])

You can simulate an SCR with an NPN and PNP transistor (see attached)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Post Edited (Beau Schwabe (Parallax)) : 2/17/2009 2:22:23 AM GMT

mctrivia · 2009-02-17 02:05

how does that otp2 work?

i tried the otp diagram. a being propplug tx, b being prop rx, d grounded, 0 ohm r1 with 10k between c & d. works for programing. applying 6v unregulated to a & c (removing prop plug) does not blow 250mA fuse. but does dim power to rest of the board.

what is the purpose of r3?

Beau Schwabe · 2009-02-17 02:16

mctrivia,

You can't use OTP for data, only for supplying power... I posted that design before you had mentioned you wanted to control the data.· Using the Prop plug in the way that you describe could damage the prop plug.

With OTP2, if the fuse is intact, then the NPN transistor is allowed to conduct.. because of the reverse biased diode, data can flow in both directions.
When the fuse is blown, the NPN transistor will no longer conduct... what was R3, now a 10K resistor .... ensures that when the NPN transistor is off, that the data on the right side stays at a logic HIGH not allowing the data on the left side to pass through.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

mctrivia · 2009-02-17 02:22

i removed the prop plug before trying to break the fuse. Can 3.3V be used with this circuit? I am not using 5v any wear in my design and this circuit requires the break power supply to be attached during normal operations.

Beau Schwabe · 2009-02-17 02:28

mctrivia,

"Can 3.3V be used with this circuit?" - absolutely!... just make sure you have enough oomph to pop the fuse... also find the lowest current fuse you can find, the NPN only needs a few milliamps to stay "on"

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Propability · 2009-02-17 02:32

Maybe this?

mctrivia · 2009-02-17 02:39

Propability that will not work because it can be bipased just by shorting the 2 terminals. the circuit needs to be such that the data can not go through by manipulation of the pins available. otp2 is good because there are only 4 access oints 3.3v,blow, data in, data out. once blown there is no way to get data flowing again.

my power supply is 1A so should be enough to blow the 250mA fuse which is the smallest i can find in 0603 size. still looking for reasonably priced in other formats.

mctrivia · 2009-02-17 02:47

If i am not mistaken if tx from prolpug is placed on the right data and rx to prop is placed on left data then the diode is not needed because this is not bidirectional communication?

Cluso99 · 2009-02-17 02:50

What about using a reed relay in the line(s) and you can energise with a magnet? No-one would know you needed a magnet to energise (enable the wire)

Except us on the forum

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:

· Home of the MultiBladeProps (SixBladeProp)
· Prop Tools under Development or Completed (Index)
· Emulators (Micros eg Altair, and Terminals eg VT100) - index
· Search the Propeller forums (via Google)

My cruising website is: ·www.bluemagic.biz

Beau Schwabe · 2009-02-17 02:53

mctrivia,

"If i am not mistaken if tx from prolpug is placed on the right data and rx to prop is placed on left data then the diode is not needed because this is not bidirectional communication?" - true but it provides a stronger drive as opposed to the weak 10K pull-up you would have without the diode.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

Beau Schwabe · 2009-02-17 02:54

Cluso99,

nice!, just put it in place of the SCR.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

mctrivia · 2009-02-17 02:57

i would like it that not even i can bypass the security measures. plus i need all super small surface mount components.

I see no flaws in Beau Schwabe Idea other then it needs about $3 worth of parts. Current draw will be minimal to almost non-existant after the fuse is blown.

Beau Schwabe · 2009-02-17 03:11

mctrivia,

Mouser has some 50mA fuses for $0.75 ... www.mouser.com/Search/ProductDetail.aspx?qs=sGAEpiMZZMseCiJT91fwIpRk8cYmTOZAWbJnmZo3urc%3d

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe

IC Layout Engineer
Parallax, Inc.

mctrivia · 2009-02-17 03:14

those are not surface mount and are way biger then the 0603 pack i am using.

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=478-2858-1-ND

heater · 2009-02-17 03:23

Connect your diode from rx pin to ground, such a way around that it does not conduct in normal use. Put your fuse from rx pin to your input.

Now a negative voltage applied to your input should conduct through the diode to ground and blow the fuse. No damage to the prop and no way for you to bypass it.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

mctrivia · 2009-02-17 03:24

nice and requires much less parts.

heater · 2009-02-17 03:26

That's odd. I just edited that post to add bold to "negative" and there is no indication of an edit appearing on it.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.

mctrivia · 2009-02-17 03:28

will the prop pin have any problems with a negative voltage or should a diode be put in to protect a negative voltage reaching the pin?

security

Comments