Help With Lock Circuit - Transistors

[Reuben98]

Circuit1.jpg is the alarm portion of the circuit. The top circuit is a turn off after delay that will turn off the signal to the alarm portion (goes through nor gate with reed switch to tell if door is open). The bottom right part is a turn on after delay which will turn the alarm on after a certain time delay. The thyristor portion in the middle has been giving me trouble. Initially I couldn't get it to work at all, but now it sort of works. The signal in the side of the thyristor sets it so even when the door is closed or opened again, the alarm will still count down/go off. The second portion of the circuit (circuit2.jpg) is a decade counter configured as a combination lock. Whenever the code is entered correctly then it should pulse then invert to a 0 so that it will turn the thyristor off, resetting the alarm etc. Unfortunately, my efforts to do so with mosfets and transistors have failed so far. Is there a way to do this, or am I going to have to use a relay (don't want to!) Many thanks, Reuben!

 

[Reuben98]

I only got the thyristor part to work when using a push to break switch, but I need it to do that using a signal from Pin 3 of the decade counter.

 

[Arouse1973]

You can only turn off the thyristor by removing the power in this instance. This could be your problem. I can't see where the inverter connects to the thyristor.
Adam

 

[Reuben98]

The inverter doesn't in those photos. I had it connected to a transistor base where the push to break switch is. The push to break switch part does work. Thats what I though, because even when the transistor is off there is a small charge trickling through.

 

[Arouse1973]

Which transistor Q7? Your better off showing us the circuit that didn't work.
Adam

 

[Reuben98]

Another issue is that the transistor, even though the collector is coming from 9v is only letting 5v through the emitter. I think this is due to the fact that the inverter is putting 5v into the base. This is causing the thyristor to just go haywire and the majority of the circuit starts going crazy. It worked fine when there was 9v going into the thyristor, but the transistor is limiting it to 5v.

 

[Arouse1973]

Which transistor, you have to tell us the circuit reference.

 

[Reuben98]

Not working circuit. Q7 works fine, its the "set" transistor that locks the thyristor "on". Its the one above it.

 

[Arouse1973]

Which is Q4? It's an emitter follower so it does not amplify voltage only base current. So what voltage you put in will appear on the emitter minus the voltage drop of the base emitter diode.
Adam

 

[Reuben98]

Circuit Wizard just gives the model as "Ideal"... Its work well as a transistor in my test circuits (I mocked up both alarm portions). I tried to use a relay, but it wasn't getting enough ampage... like 18 nano amps because of the inverter of transistor.

 

[Arouse1973]

Your LED at the bottom of the Thyristor needs a current limiting resistor and you could benefit from a pull down resistor on the gate connected to the Thyristor to common 0Volts, this should ensure it turns off when you remove power.

 

[Reuben98]

Even a pull down resitor of 330 causes the circuit to flicker uncontrollably when i set the thyristor.

 

[Arouse1973]

Try a pull down on the gate of the Thyristor see if that helps. Edit: You may need a series current limiting resistor also for the gate, depends how good the simulation is.

 

[Reuben98]

That helped, and stopped the flickering of the 330 on the LED. Still trying to figure out how to replace the push to break switch though. My relay won't even work as it doesn't receive enough ampage from the transistor/inverter. It's getting late so I'm going to try and get a fresh start at it tomorrow.

 

[Arouse1973]

Ok post the relay circuit tomorrow and I'll have a look.
Adam

 

[Reuben98]

Circuit 6 works. The relay receives 15mA and so closes the switch correctly. As can be seen on circ5 it is all working, or would be if the relay received 15mA instead of a minuscule amount which won't trigger the relay.

 

[KrisBlueNZ]

Hi Reuben and welcome to Electronics Point

Your designs seem to show some understanding of circuit design but there are many errors. I'll start with the obvious ones.

First of all, you should start by redrawing the circuit with the positive supply rail along the top, and the negative (0V) rail across the bottom. This makes it a lot easier to follow and understand the design. And put the battery at one end. As far as possible, put the inputs at the left and have the logical signal flow from left to right.

Can you explain the reason for D7 and D13? It's not normal to interrupt the 0V rail like this, and I don't see any good reason for it.

Where are the ICs' VDD and VSS pins connected?

Why is the Set Alarm pushbutton connected to drive the Door Open LED?

Is SW3 the reed switch you mentioned? You need a pulldown resistor on the input to IC1A. If the switch is open, the input will float and will not assume a known state unless you have a pulldown resistor.

You need a current limiting resistor in series with D5. (Or is it D6? The one from the SCR's cathode to the 0V rail.) Adam pointed this out as well.

You definitely don't need a relay and you're fighting a losing battle with the SCR. You would be better to use two NOR gates from IC1A as a set-reset flip-flop. Easier to control and monitor. Google NOR gate SR flip-flop.


The combination lock circuit has a few problems. The idea of having an overall timeout is good, but you still have to detect incorrect button presses, otherwise you just need to press all of the buttons randomly and the lock will open. There are ways to do this and I'll look into it and suggest something.

 

[hevans1944]

I don't even want to open this can of worms now that @KrisBlueNZ has chimed in. But I will think about it overnight.

On push-button combination locks:

The best one I have ever seen was in the entrance to a highly secure area. It consisted of a plainly visible numeric keypad with blank keys and a numeric display that was inside a box you had to look directly into to see the digits. The digits were not visible by anyone standing to either side or behind you because of the way the box was constructed. Now here is the kicker: when you pressed the "START" button, the numeric displays inside the box would light up with the digits 0 through 9 in random positions. Each valid user knew his or her six-digit (or whatever) key code, but that code had to be entered on the blank keypad in the same positions as the displayed digits. The digits were scrambled each time the "START" button was pressed. It took a little getting used to, memorizing the displayed digit positions and then keying in (in proper order) your personal code.

After keying in your code you had to press an "OPEN" button which either opened an electric lock on the door or did nothing. There was no feedback to let you know if you had entered the digits correctly or even if you had entered the correct number of digits. And what was so important that this lock was guarding? A jet engine test cell.

I have seen and used a lot of "secure" entrance methods that rely on switch operation (there are more sophisticated biometric locks for really secure areas) but this little puppy really impressed me in its simplicity and ingeniousness. It could have been made even more secure by hiding the keypad inside a shroud, so only haptic or tactile senses could be used to identify which keys to press. That was probably a bit too much bother though, except maybe for the blind, who wouldn't be able to read the numeric display anyway. A keypad with programmable Braille readouts that could be scrambled would work fine for blind users and could be a marketable product.

 

[KrisBlueNZ]

That's very cool, Hop. I had trouble understanding the description until I realised that the numbers that were lighting up weren't actually on the keys; they were on a completely separate display that you had to put your head up to. Then you pressed the unmarked keys in the positions corresponding to the numbers in your combination, according to the positions shown on the display. Is that right?

You said "memorising the displayed digit positions and then keying in ...". Did the display only light up for a short time, so you had to remember it?

I like the idea of having a Start button and an Open button, and the device doesn't give you any clues at all. You have to know exactly what you're doing. If you do, it's straightforward enough. If you don't, you are simply ignored.

 

[hevans1944]

... Then you pressed the unmarked keys in the positions corresponding to the numbers in your combination, according to the positions shown on the display. Is that right?

Exactly right!

You said "memorising the displayed digit positions and then keying in ...". Did the display only light up for a short time, so you had to remember it?

The display remained lit, but it was somewhat difficult to stare at it and manipulate the keypad buttons simultaneously. Most users would "find" the positions of their numbers on the display and then look away to the keypad to punch in those numbers, perhaps looking back to the display if they had a longish code sequence. Alas, I don't remember ever being authorized to enter the area without an escort, so I don't remember being assigned a code to use. I think for this level of "security" they would use a four, five, or perhaps six digit code and allow duplicate digits in the sequence. I have never seen this particular lock used before or since I saw it one time while on a service call there. I had the proper clearance level, but there is also "need to know" that requires documentation and approval... so this customer, since it was supposed to be "one time" visit, didn't bother giving me a code. They provided an escort instead.

I like the idea of having a Start button and an Open button, and the device doesn't give you any clues at all. You have to know exactly what you're doing. If you do, it's straightforward enough. If you don't, you are simply ignored.

Exactly right! It is common at some "secure" entrances to use a panel, somewhat shrouded, with four or five bi-directional rocker switches to enter your code. Protocol says you don't let anyone see what rockers you press, but this IMHO is not proper "security". The rockers are quick to use though and more convenient at high-traffic entrances. We used them at my place of work in the lobby to gain entrance to the rest of the building, but everyone had the same code! To change the code you had to re-position jumper wires in the lock on the "secure" side of the entrance. Pure crap IMHO, but it limited access to strangers... at least until they surreptitiously observed you entering the code.

With the separate display/keypad approach, everyone could have their own code and entrance could be logged and time-stamped. To be truly effective building security, there needs to be a display/keypad for exit too. The facility I visited only used entrance access control. Once you were inside, you were free to leave at any time without needing to enter a code because there was no place provided to enter a code!

Of course the best security is a human being who recognizes the person seeking to gain entry.

One of the places I worked at as a civilian contractor (Foreign Technology Division at WPAFB) used an entrance door that anyone could enter but the door locked behind you, trapping you in a small area where your face and photo ID credentials were checked by a real person. If everything was in order, the next door unlocked and you were free to go about your business. If things were NOT in order, you waited for armed security to arrive to escort you somewhere else. There were several of these "entrance locks" in the building since everything was "compartmentalized" for security. You could work for years there without knowing what your co-worker down the hall (in a different compartment) was doing. These folks were (and still are, under a new name) really serious about secrets. It took almost eight years and a classified briefing before I was granted unescorted "access" to this facility.