Society of Robots - Robot Forum

Mechanics and Construction => Mechanics and Construction => Topic started by: gamefreak on February 22, 2007, 03:50:25 PM

Title: RFID Door lock
Post by: gamefreak on February 22, 2007, 03:50:25 PM
hello all.
I am currently working on a RFID door lock and so far my current idea for the lock is a solonoid.

But is it possible for the solonoid to move in to let the door close?
Also if i use a normally open solonoid can i push the solonoid arm in without damaging it?
Title: Re: RFID Door lock
Post by: ed1380 on February 22, 2007, 08:23:47 PM
Yes you can move the rod. Just bevel the rod so that it looks like the rod on a door handle.
Title: Re: RFID Door lock
Post by: gamefreak on February 22, 2007, 08:40:04 PM
Now how do i bevel something? Just sand/cut it to the shape?

Also a normally open solenoid is a pull type right?
What is the stroke? i found this website:http://www.solenoids.com/linear_solenoids.html
But what does the stroke mean?
Title: Re: RFID Door lock
Post by: gamefreak on February 23, 2007, 11:48:50 AM
Another solonoid i was looking at it:http://solenoid.com/images/solpdf/st1325n.pdf

But im a little confused on the Duty thing. 100% has the most resistance least force and least wattage(lost?) and 10% has most force least resistance and most wattage(lost?)

So for a door i probably want 100% duty because i dont need force and it lets off the least heat?
Title: Re: RFID Door lock
Post by: ed1380 on February 23, 2007, 12:14:31 PM
Is it a magnetic silinoid that will pull in the rod to let the door open or one where the rod is threaded and a motor pulls it in by spinning
Title: Re: RFID Door lock
Post by: gamefreak on February 23, 2007, 12:43:03 PM
i think a magnetic because it is simplier by when i get a valid card i apply the voltage and it will unlock the door.
Title: Re: RFID Door lock
Post by: ed1380 on February 23, 2007, 03:05:15 PM
You could take an existing door handle setup, and replace the handles with a servo.
Title: Re: RFID Door lock
Post by: gamefreak on February 24, 2007, 04:35:59 PM
i thought about that but i cant figure out how i would hook it up.

Any ideas?
Title: Re: RFID Door lock
Post by: ed1380 on February 24, 2007, 05:51:12 PM
or maybe a deadbolt. just remove the deadbolt handle, and ghetto a servo on. It would depend on what servo and what door handle it is.
Title: Re: RFID Door lock
Post by: Admin on February 24, 2007, 05:54:37 PM
a servo driving a deadbolt:
http://cobweb.ecn.purdue.edu/~vnhhs/Archive/fnl_rpt_s9a8.html

(http://cobweb.ecn.purdue.edu/~vnhhs/Archive/Lock.JPG)
Title: Re: RFID Door lock
Post by: gamefreak on February 26, 2007, 08:10:29 AM
The servo plasted on to the latch would work, but it doesnt look very good, i'd rather have it all hidden inside the door or not have it extruding those extra two inches by attaching the servo.


Also if i had it go back to locked before the door closed then the door would not lock shut.
Title: Re: RFID Door lock
Post by: Admin on February 26, 2007, 10:07:56 AM
the CMU robotics club used to have a servo driven door lock similar to this. we had a big red button you can push to manually unlock/lock the door, plus a magnetic card swipe outside that unlocks two doors for a short time period, and logs your visit.
Title: Re: RFID Door lock
Post by: gamefreak on February 27, 2007, 09:35:41 AM
I geuss i'll change my overall plan so that it could be a lock for a safe or something and not to get into or out of a building, my reader and card arrived yesterday from parallax and i hooked it up so that first scan will open the servo and second scan will close it, but since its a continuos rotation if my progect runs out of power halfway through rotation then it will be messed up
Title: Re: RFID Door lock
Post by: ed1380 on February 27, 2007, 02:44:37 PM
i don't know if this will work in your situation, but maybe have the deadbolt have a spring pulling it locked, and the servo works against the spring to open it. if power fails the spring should be strong enough to close the deadbolt. but if the door is open atm then that wont be good.
Title: Re: RFID Door lock
Post by: gamefreak on February 27, 2007, 02:54:26 PM
So far i seems that my original idea(the solenoid) would be the best, but the ony thing i know is that i need a pull type that is small enough to fit inside of the door or the door frame. I have a "Parallax Board of Education, USB" and it gives up to 5 volts regulated, so i know i cant exceed that, another thing is that this will probably have to be battery powered so i need to have high current batteries(i think) in order for this to even be remotly possible without the batteries dieing after one use.

If someone could explain Stroke and Duty Cycle to me then i could get solenoids to work, also if i could get a link to batteries that might handle say 100 uses with the:BS2 "http://www.parallax.com/detail.asp?product_id=BS2-IC", RFID Reader Module "http://www.parallax.com/dl/docs/prod/audiovis/RFID-Reader-v1.2.pdf", and the solenoid, then i might be in buisness.

One solenoid i ran across at http://w2s.solenoids.com/ledx/ds/LXEZR/LXEZR2e.lasso?pcode=L214 is the .5 by .5 STA Pull has a speed of 4 milliseconds to move .5 millimeters with 100 percent duty(i think always on) and a force of 4 OZ at about 1 millimeter

Am i understanding that right because that sounds very small and short, but it might work for a door prototype....
Title: Re: RFID Door lock
Post by: JesseWelling on February 28, 2007, 03:19:54 AM
what about chain driven from a servo that's embedded into the door?
Title: Re: RFID Door lock
Post by: gamefreak on February 28, 2007, 06:39:45 AM
Doors are only about an inch thick and servos are an inch and a quater or an inch in a half depending where you measure from, so you would have the horn stiking out, which might be helpful, but with a chain anyone could just break it or something, Also the point is not to unlock the door, but to unlock and lock it in a way that you can only do if you have card, if they broke the chain or forced it to turn then anyone could get in.
Title: Re: RFID Door lock
Post by: Admin on February 28, 2007, 10:54:25 AM
Quote
One solenoid i ran across at http://w2s.solenoids.com/ledx/ds/LXEZR/LXEZR2e.lasso?pcode=L214 is the .5 by .5 STA Pull has a speed of 4 milliseconds to move .5 millimeters with 100 percent duty(i think always on) and a force of 4 OZ at about 1 millimeter

Am i understanding that right because that sounds very small and short, but it might work for a door prototype....
Solenoids tend to have a small travel distance . . . you will have to shop around for awhile. My vote is for the servo on deadbolt idea :P

As for your power concerns . . . you can have both plug in power, but also battery backups in case power goes out. Also, you can put a timer on it, so say if nothing happens for an hour, it will reset itself (look up watchdog timers), and relock the door. For something like this, you want backups.

Also, for the robotics club door, after the system locks the door, the servo turns off. So if say the system fails, you can still open the door with a normal key.

Multiple backups are good :P
Title: Re: RFID Door lock
Post by: gamefreak on February 28, 2007, 04:36:47 PM
so your saying to buy a normal door handle with a lock, remove one side, attach the servo to that side, and then it should work?
Title: Re: RFID Door lock
Post by: Admin on February 28, 2007, 07:54:52 PM
The servo rotated the deadlock. Just attach the servo to the part you rotate to lock the door. May or may not be the best method, but the method I mentioned worked very reliably.

Oh, we also put a small buzzer on it that buzzed when you swiped the card. Good feedback to know its working properly.
Title: Re: RFID Door lock
Post by: gamefreak on March 02, 2007, 11:48:07 AM
Alright, as soon as my RFID reader rearrives then I can then set up the deadbolt and servo, Since i have free reign of all the machines and scrapwood in the woodshop then i can build a small door and a holder for the servo and deadbolt.

Now all i need is a battery....
Title: Re: RFID Door lock
Post by: dunk on March 02, 2007, 02:16:57 PM
so all the way through this you have been talking about putting the actuator inside the door.
have you considered putting it inside the door frame?
it would allow for a far bigger mechanism.

dunk.
Title: Re: RFID Door lock
Post by: gamefreak on March 02, 2007, 04:06:53 PM
actaully yes, when i first saw the door latch pictue that was my idea, so far my shop teacher and I are trying to figure out the placement of everything.


Alright now for a battery... I was thinking of a Li-po, Li-Ion, or Nicad battery because of there higher current and the power consumption of my circuit, would any of these work?
Title: Re: RFID Door lock
Post by: Admin on March 05, 2007, 09:00:50 PM
you probably would want a NiHM type so its long lasting, cheap, and easy to recharge . . .
Title: Re: RFID Door lock
Post by: gamefreak on March 06, 2007, 10:25:07 AM
Now on the links i gave for the Basic stamp and RFID reader, the power consumption is 93 MAh i think, so that means with a battery of 2300 MAH i can last 24 hours

Am i right?
Title: Re: RFID Door lock
Post by: Admin on March 06, 2007, 10:55:31 AM
93mAh would be totally incorrect if you also wanted to power the servo with the same battery . . .

but yea, 2300mAh/93mAh = 24.7 hours

does the basic stamp have sleep functions (low power draw do nothing but wait mode)?
Title: Re: RFID Door lock
Post by: gamefreak on March 06, 2007, 11:42:25 AM
Yes i know i didnt count the servo but since i dont expect it to run all them time im not to worried about its power draw.

Yes the basic stamp does and so does the RFID reader if i but them both in sleep mode then.... i would use 10.5 MaH and then bring up my battery life to almost 21 days,
the basic stamps leep mood is 5 uA and the modle is 10 mA

I dont know how much current the servo uses because the spreadsheet only gives me the volatge it can handle.


One thing i could do is say every two second turn off the RFID reader module and then turn it back on two seconds later and that should add more life to the batter due to the fact that im taking 80MaH out of the usage for 2 seconds

Due to the fact the the board has a jack for power other then a 9 volt battery and i dont see any batter packs with a jack i will have to attach my current battery choice of 8.4 volts and 2600 mAh to it, but is this good for the electronics or will it not make a difference?
Title: Re: RFID Door lock
Post by: dunk on March 06, 2007, 03:08:33 PM
so servos use quite a lot of power whether they are moving or not.
you can get round this by switching on the power to the servo using a MOSFET only when needed.

i haven't actually played with sleep mode on any microcontrollers but i think you wake them using an enable pin.
what about your RFID reader? can it detect when a card is in front of it while in sleep mode or will it need you to press a button?
if it can detect a card from sleep mode maybe it has an output you can use to wake your basic stamp when it does.

which board has the 9V jack? i'm guessing that's the RFID reader right?
i'd imagine 8.4 volts should work ok.

your basic stamp and servo will want around 5 Volts.
if you want maximum battery life i'd recommend a switching power regulator to drop your 8.4V to 5V.
you can build one using a LM2574 pretty much straight from the datasheet found here:
http://www.national.com/pf/LM/LM2574.html (http://www.national.com/pf/LM/LM2574.html)
switching step down voltage regulators are far more efficient than other types.

happy building!

dunk.
Title: Re: RFID Door lock
Post by: gamefreak on March 06, 2007, 04:54:32 PM
my board with the basic stamp on it, its called the Board of Education and its from www.parallax.com

No the RFID reader can only read when it is in active mode so the only way to save energy is swithing it on and off.
Im not sure about the basic stamp and i dont have any MOSFETs, if power becomes a major problem i'll go to radioshack and buy some to enable and disable the servo
Title: Re: RFID Door lock
Post by: Admin on March 06, 2007, 10:19:59 PM
another place to buy a switching regulator
http://www.dimensionengineering.com/DE-SWADJ.htm

but the problem with any regulator is really low efficiency so you should avoid it if you can
Title: Re: RFID Door lock
Post by: gamefreak on March 07, 2007, 04:17:54 PM
i dont think i'll ned a regulator, the BOE has a built in regulator and 3 VIN pins so i can choose between regulated and un-regulated. Also if my batterys dont holf a charge long enough i can figure out how to use MOSFETs and hook up the servo to it
Title: Re: RFID Door lock
Post by: gamefreak on March 30, 2007, 11:54:11 AM
Alright i bought a door latch sanded it down a bit to make it flat and epoxied a servo onto it, but im using http://www.parallax.com/detail.asp?product_id=900-00008 and im not sure if its enough torque, i came to the conclussion that if it isnt then i'll modify the code to make it turn just enough to move the latch but so that it doesnt go into the indents that keep the latch from beign pryed open

heres a bad diagram of what i mean, the latch has two indents(indicated by the gaps) that stop it from being forced shut, it requires a little bit extra torque to start the unlocking but for the rest it is pretty minimal, so i figure if it cant get the latch out of the indents then dont let it, will this work?
 _ ______ _
| ||         || |
Title: Re: RFID Door lock
Post by: Admin on March 31, 2007, 02:08:19 PM
haha you diagram was too bad, i cant picture it :P

what is resisting the rotating servo? perhaps you can grease up the joints or make the moving lock thingy a loose fit?

the best way to test required torque is to get a stick of X length, attach it to the part on the lock that rotates, then add weights to the end of the stick. keep increasing the weight till the lock turns.

weight * X = minimum required torque
Title: Re: RFID Door lock
Post by: gamefreak on April 01, 2007, 09:41:56 AM
Well i epxoyied the servo on and i geuss i got lucky, i preprogrammed the continous rotation servo to turn about 180 degrees, if i hold the servo and latch the it turns just enough to lock/unlock the latch if its not in the ditches

Also i bought a 8.4 volt NIMH battery pack today and it has a "Deans Ultra" connection, do i need a special charge for those or will any old NiMH charger work?
and to hook it up to the BOE, from parralax should i just take two jumper wires and connect it to the 9v battery spot? sorry for all the questions.

here are the ditchs/gaps im talking about:
(http://i9.photobucket.com/albums/a90/monkey4sale/IMG_0437.jpg)
(http://i9.photobucket.com/albums/a90/monkey4sale/IMG_0436.jpg)
Title: Re: RFID Door lock
Post by: Togo5 on April 11, 2007, 04:15:01 PM
i once made a lock and i thought it worked quite well.

i made an electromangnet using a soft iron core, then attatched it onto the door. in line with it (the 'latch') i used a magnetic rod. to lock the door i switched the + - to make the magnet repel the object thus moving the latch right into the stopper locking the door.

to open the door is just simply swiching the leads again to use it as a normal electro magent with it attracting the rod and freeing the door

just thought you might wanna hear other possibly ideas  i think it worked really well for such a simple idea. :)
Title: Re: RFID Door lock
Post by: zamboniman60 on April 14, 2007, 03:03:30 PM
If you're going to be using this in a stationary application, like pemanently attached to a door, wouldn't it be a good idea to just find the right wall transformer/rectifier/filter box and just run a long wire?
Title: Re: RFID Door lock
Post by: Admin on April 24, 2007, 06:16:14 PM
Quote
Also i bought a 8.4 volt NIMH battery pack today and it has a "Deans Ultra" connection, do i need a special charge for those or will any old NiMH charger work?
I agree with zamboniman60's suggestion of using a plugable power supply . . . anyway, 8.4V is a little odd, but any NiMH charger that says it can do 8.4V will work fine. but for the connector . . .

the Deans Ultra connector is really odd looking
(http://www.siriuselectronics.com/images/Deans%20Ultra.jpg)

you can always solder some wires on to it, or cut it off and change the adaptor . . .

Quote
and to hook it up to the BOE, from parralax should i just take two jumper wires and connect it to the 9v battery spot? sorry for all the questions.
you mean connect the 8.4V battery to the power connection labeled 9V on the boe bot? if so, then yes.


again, I would just get a power supply you can plug into the wall . . .