Pi driven relay

Go To Last Post
67 posts / 0 new

Pages

Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A friend of mine is driving a low current relay coil directly with a gpio pin on a Raspberry Pi.  The circuit is Pi pin to relay coil + pin.  Relay coil - pin to ground.  Does he have to worry about the back EMF blowing up the Pi pin when the voltage from the gpio pin goes from +3.3V to 0V.  Would the solution be to put a schottky diode between the Pi gpio pin and the relay coil + pin to keep the back EMF from getting into the Pi?  Or should the Pi be driving a FET?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes to everything. smiley

 

Nice to see you back MArk

 

JIm

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Jim.

The easiest thing would to be to just put a schottky between the gpio pin and the relay coil.  Will that work?  Does the Pi have a protection diode that does the same thing?  

 

If I drive a FET gate with the gpio pin, that circuit is usually 3.3V -> coil -> FET drain, FET source grounded, with a diode from the drain back up to the 3.3V supply.  The problem there would be the 3.3V would be coming from the Pi 3.3V pin, and then the back EMF would be shunted back into the Pi, which doesn't sound good either.

 

Nice to be back,

mark

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Or in the first scenario is the schottky going to break down under a back EMF? 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The Pi is Vcc is 5V, and it has a 3.3V regulator with a 3.3V output pin where the 3.3V would be coming from in the FET scenario.  The relay is 3V with a 150 Ohm coil.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Not sure of the Pi port pin specs, but the relay will pull around 20mA - that might be over the edge for the Pi - best to check. Yes, you will need a diode across the coil - a 1N4148 would suffice.

 

I'd suggest using the many cheapy relay boards available - there seems to be a few varieties with some being low active and others being high active as they have a transistor on them. The high active one is the one you'd want. 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Drive an N fet gate through a resistor from the pi (say 1K).  Add a 5V zener from the gate to gnd & also gnd the fet source...the optional zener protects the gate from esd.   Tie the relay coil to 5V (or 12v) & the other end of the coil to the fet drain.  for small relays no diode is needed, the fet 's internal zener will protect (assuming a decent current rated fet).  If desired, a diode can be used on across the relay, though it will slow down the coil shutoff slightly (like ms times).

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yeah, we may need to put a couple of gpio pins in parallel to get the 22 mA the relay data sheet quotes.  From what I have found the default current from a Pi gpio pin is 8 mA, but you can set something in the Pi to get 16.  That is the problem of the guy I am helping. 

 

Can I just put a diode across the coil and be done with it?  That seems like it feeds the back EMF back into the gpio pin, or do I have the directions reversed.  I was thinking I could put a schottky in series with the coil, but realized the reverse breakdown voltage is much lower than the back EMF.

 

He already has the relay.  It is 3V, 22mA.  $9 from Digikey.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yeah, we may need to put a couple of gpio pins in parallel to get the 22 mA the relay data sheet quotes.  

 

Well that might be ok, but has trouble spots...the relay may take a much larger surge than 22 mA during turn on, and unlike the tough AVR, the pi IO seems somewhat unforgiving & easy to blow out.   If the io aren't balanced, one io might bear too much of the load, give up the ghost & form a cascade of  trouble.   Why not add a 10 cent fet & save 2 IO?  Even a 2N2222 is "ok" (though you get Vloss & then need a diode on the coil).

 

You will NOT be putting a diode between the pi & coil under any configuration 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
Drive an N fet gate through a resistor from the pi (say 1K).  Add a 5V zener from the gate to gnd & also gnd the fet source...the optional zener protects the gate from esd.   Tie the relay coil to 5V (or 12v) & the other end of the coil to the fet drain.  for small relays no diode is needed, the fet 's internal zener will protect (assuming a decent current rated fet).  If desired, a diode can be used on across the relay, though it will slow down the coil shutoff slightly (like ms times).

 

Yes, I have done this many times making stepper motor controllers, with the protection diode pointing toward the voltage source across the coil to protect the FET.  In this scenario, the voltage source is the gpio pin driving the relay, and I want to protect that pin without introducing a FET if I can.  If I cant, then the voltage source in your scenario comes from a 3.3V regulated pin on the Pi, and then I worry about blowing it up.  Everything has to come from the Pi.  Does that make sense?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How does the diode across the coil put stuff back into the port pin? The reverse current circulates in the coil and diode. I too suggest a transistor/mosfet. The gpio on the Pi was never designed for real world action - it was meant for a set top box.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
Why not add a 10 cent fet & save 2 IO?

You are probably right.  I need to find how much current the 3.3V output pins can supply.  The total current from all the gpio pins cant add up to more than 51 mA, and putting them in parallel is probably a bad idea.  The Pi datasheet is a joke.

 

The protection diode has always been a little mysterious to me in the circuit.  It protects the FET turning the current through the coil on and off, but what protects the voltage source driving the coil.  With the direction of the diode across the coil it sends the back EMF into the voltage source.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

if you're talking about the intrinsic diode in a mosfet, then, yes, the current does get shunted into the supply rails. The idea is that the impedance of the supply is low. With the diode ACROSS the coil, the current flows in a loop.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
The reverse current circulates in the coil and diode.

I guess that answers my question.  Does a large voltage never appear where the coil is connected to the voltage source? 

 

I think I have been overthinking it.  What I remember from my E&M classes is the induced EMF is always in the direction such that it creates a magnetic field in the direction opposing the change in changing magnetic field.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
if you're talking about the intrinsic diode in a mosfet, then, yes, the current does get shunted into the supply rails. The idea is that the impedance of the supply is low. With the diode ACROSS the coil, the current flows in a loop.

 

In that case, if I drive the relay with 16mA from one gpio pin and put a diode across the coil, then I dont have to worry about blowing up the gpio pin.  Is that right?  

 

We ran the thing with whatever default current came out of the Pi and it drove the relay and we saw the resistance across the switched pins go from very large to very small with no voltage being switched and no load, like it was working as intended.  We ran out of time and couldn't measure the coil current.

 

The idea of putting a diode in series with the coil was just a dumb idea of mine before I thought about it.  Now it makes no sense and is embarrassing.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

OK, the gpio pins are wimpy, so as pointed out, the surge current could destroy things.  It is not a mega.  After a little searching, the 3.3V power pins can put out 700ma, or something huge like that, and the 5V pins are connected to the input mains, so they can supply whatever you want.  So the solution is to drive the relay coil with the 3.3V power pins, and switch the relay current with a FET and a protection diode just for fun.  Thanks for the help guys.  I appreciate it.

Last Edited: Thu. Mar 12, 2020 - 05:53 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
the relay may take a much larger surge than 22 mA during turn on

Wouldn't the inductance of the coil reduce the surge compared to a straight wire?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

good call in #23.

 

An aside explanation:

The internal fet zener "trips" when the coil is shut off, since doing so produces a high positive voltage at the drain. 

Why? The inductor produces a voltage proportional to how fast you turn it off  (Vcoil= L* deltaI/deltat).  You turn it off fast with the fet, hence a large NEGATIVE voltage instantly develops (deltaI is negative (decreasing current), hence so is VL).

Now you already set things up defining 5V (or 3V) where the top  is positive and the bottom (at the fet) is negative for your positive 3/5V. Now we come along & say we have a negative coil voltage, so that means the fet terminal is higher than the 3/5V supply. 

This voltage tries to rise really high (it is trying to go as high as needed to force the original current flow, so it might create 1000's of volts trying), except upon hitting the mosfet zener trip voltage, the FET zener kicks in and conducts the original current to gnd (making the coil happy).  Of course since the driver is off, the energy quickly fades out.   

 

Say you have a 30V rated fet, its zener will kick in around 30V.  If your supply is 5V, that means the coil will have 25V across it during the initial turn off.

If you add a diode across the coil the voltage across the coil will instead  be about 0.7V during turn initial off.

However, rearranging you see the turn off time: deltat = L*deltaI/Vcoil...a larger coil voltage gives a faster turn off time (smaller deltat)...the 25V coil voltage gives a much faster delta I turn off time than a 0.7V coil.   This is most useful for valves or when you need the fastest "immediate" reactions.

 

If the fet is "weak" (or you use bipolar NPN)  or you just really want to use a diode on the coil, you can instead use a series diode-zener combo to get the same voltage improvement effect...put them  back-to-back series across the coil, with the zener pointing towards the FET/NPN.

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Thu. Mar 12, 2020 - 06:25 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

Wouldn't the inductance of the coil reduce the surge compared to a straight wire?

 

If it were just a pure inductor, yes, but this is a relay with moving parts.  It really depends on the specific, size of the relay, the amount of mechanical mass moving around, etc...some relays are HUGE.

 

 Just as in the coil of a solenoid, the amount of current that the coil uses when it's first energized and pulls in will be larger than the hold-in current. The pull-in current may also be called the inrush current and it's a rating that's normally listed in the relay specifications. The inrush current is generally three to five times larger than the hold-in current, which may be called the seal-in or sealed current, and it will also be listed in the specification for the coil. The diagram for pull-in and hold-in current for a relay coil is similar to the example of pull-in and hold-in current for solenoids shown previously (see below).

 

 

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Thu. Mar 12, 2020 - 06:26 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I get easily confused with a written description of a circuit.

A picture is clearly worth a 1000 words!

 

The left image shows a relay being driven by a "low side driver" transistor, with a reverse biased diode across its coil.

 

Side comment:

A small piezo doesn't need a series cap or a series resistor...

 

The NFet's  Gate pull-down resistor, above, is to hold the NFet off during the micro's power up, when an AVR's pin is in Input mode, and hence the Gate is floating.

One wants the NFet to be either switched On or Off, and not in some partly turned on analog state in the middle, where it's Rds is higher, and it can over heat.

 

Next:  The 2N7002, shown above, is a great little NFet for 5V circuits.

 

I'm not convinced it is so great for a 3 V circuit.

I'm not sure it turns on as fully as desired with a 3 V gate voltage.

 

It would be interesting to hear of some other low power NFets with a low Gate turn on voltage used by others in their 3V circuits.

 

Another side comment:

If your circuit has both 3V and 5V power rails available, you might well power the relay from the 5V rail.

That would limit the current drawn by the 3 V regulator, if it happens to be rather limited either in current capacity, or in its heat sinking on the board.

(That assumes that the 5V supply is "heftier".)

That doesn't, however, change the fact that you still have a 3V pin driving the NFet's Gate.

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks candy, that helps a lot.  I was thinking I would use a 2N700 FET as they are small and I have them.  The datasheet shows a diode from source to drain, but no zener.  And probably a 1N4148 protection diode as I have hundreds of them.  It is a small relay and switching speed is not important.  It is for a door lock project my friend is doing for the post office here.

 

Kartman wrote:
How does the diode across the coil put stuff back into the port pin?

I keep thinking about the huge voltage instantly developing at the top of the coil that is connected to the gpio pin.  I am forgetting that if the diode is there it is going to "instantly" conduct and the current will circulate in the coil dissipating the voltage, so to speak.

 

I am totally overthinking this and am off in la la land of self inductance and cant see the forest for the trees.  You'all know how I am.  V = L dI/dt says it all.

 

avrcandies wrote:
good call in #23.

23?

 

Thanks guys.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hey Doc,

You and I must have been typing at the same time.  I have used the 2N7002, which is I believe the surface mount version of the 2N7000.  I also seem to remember problems with it turning all the way on with 3V.  The TO92 has worked OK for me, and the datasheet shows a max gate threshold of 3V for both, so it is possible to get a "bad" one.  The Pi gpio pins put out 3.3 V.  The relay we are using is rated for 3V, 22 mA, and I'm sure could be run at 5V for short times no problemo.   Getting at the power pins may be an issue.  He has an LCD plugged into the top of the Pi, that covers up the bank of gpio and power pins.

 

What I got from Google is the Pi 5V pin is connected directly to input power, and the 3,3V pins are connected directly to the voltage regulator driving the ARM and everything at 3.3V.  I think it was a 1117 type low dropout regulator with 700 or 800 mA available. 

 

Yes, we will be using the circuit on the left side of your post.  Connecting the relay directly to the gpio pin with no diode was a beginners mistake. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

 The datasheet shows a diode from source to drain, but no zener. 

They are actually lying, it is a "zener" with a breakdown typically, at the FET max VDS rating.  Some mfgrs show it in the symbol, which is more informative & accurate.  Many hope you won't know about the zener effect, and try to brush it under the rug, none the wiser. In higher power situations, these are characterized, optimized, guaranteed to take a beating (called avalanche rated fets).

  

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Fri. Mar 13, 2020 - 07:24 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


The 2N7000 datasheet shows a regular diode I have often wondered about.  Thanks so much for explaining that!   BTW, my friends Pi is not fried.  I am getting together with him tomorrow to add the FET and diode to the little circuit board we made.  Getting to the power pins on the Pi is going to be a kluge.  He said the LCD uses all the power pins.  Maybe the 5V pin will be available.  I would run that 3V relay at 5V without much worry.  I guess I could put in a 100 Ohm resistor to drop a couple volts at 22 mA.  I feel like such a dufus for forgetting about the back EMF from relays.  I dont use them very often.  O well, I learned something so that makes it a good day.  Thanks again.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Be very careful...on the pi the 5V & 3.3V pins are next to each other...the slightest momentary short & the pi is thereafter quickly rendered into the "as-is" sellers bin...and the is, is not.

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

OK, thanks for the tip.  I guess it makes sense that the output of the 3.3V regulator goes everywhere and the ARM is not 5V tolerant.  I used to be cocky and would work on a circuit that was still powered up, being too lazy to unplug power.  I dont do that so much anymore.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

OK, everything is working without a protection diode across the relay, which was not intentional but because I forgot to solder it in.  After all of this my friend decided it would be best to get relay power from the 12V the relay is switching instead of the PI.  So I soldered in the FET and 3.3V regulator and promptly forgot the protection diode.  I guess candy is right that the diode inside the FET works to protect things as it works fine. 

 

I am going to add a protection diode anyway because that's the way I roll.  The relay is switching a door lock which also has a coil, so I assume I need to put a diode across the 12V going to the door lock to protect the points inside the relay.  I never imagined a relay would be such a learning experience.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No protection diode does not mean instant death - usually it is death by a thousand cuts.

 

The relay arcing is the least of the problems when there is no back EMF protection - the transient is rather significant and can upset the Pi quite easily. The diode for the lock coil needs to be placed at the lock coil, not at the other end. You be surprised at what mayhem the transient can create.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
The diode for the lock coil needs to be placed at the lock coil, not at the other end.

If it is the standard security mag lock that is common here in the states that is not so easily done as the access pocket is quite small, and the connector is not positioned to do this as well.

 

Most access control vendors usually supply MOV's to place across the terminals of teh lock relays to suppress the back EMF and I have systems out there that have been going 20 years with no issues using this setup.  Honeywell sells a snubber network that you must purchase(cheap bastards) to place across the terminals at the panel.  Does the trick as well...suppresses teh spikes, and drains teh wallet all at the same time.

 

JIm

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Movs are probably ‘softer’ than diodes and allow a higher voltage than a diode. The wiring between the relay and the lock form a big loop that can radiate, so minimizing the loop is advisable.

Last Edited: Tue. Mar 17, 2020 - 12:35 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why the h*ll would you use a mosfet with 7 OHMS of resistance?  There's about a billion fets to choose from nowadays, and even 0.1 ohm is now considered fairly bottom of the barrel.

Even if you are driving only 50ma, that's a huge voltage drop.

 

exactly what relay are you driving?

 

 

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:

Why the h*ll would you use a mosfet with 7 OHMS of resistance? 

 

exactly what relay are you driving?

 

The FET is what I had that was recommended to me long ago.  This is the relay. 

 

https://www.digikey.com/product-detail/en/te-connectivity-potter-brumfield-relays/V23026A1006B201/PB2650-ND/2398017

 

Maybe because it is such a crummy FET it wont turn off fast enough to give a big dI/dt.

 

Kartman wrote:
The relay arcing is the least of the problems when there is no back EMF protection

Avrcandies is telling me I dont need the diode because the FET has a built in zener.  I am putting in the diode anyway, as I dont want to upset the Pi and make a mess.

 

jgmdesign wrote:
If it is the standard security mag lock that is common here in the states that is not so easily done as the access pocket is quite small, and the connector is not positioned to do this as well.

 

I think it is standard.  My friend got a couple of them from China, I think.  One of them came with a diode.  On the one we are using the terminals in the pocket have little tabs on them with holes to solder to.  I thought it was for soldering the bell wire, but is probably for soldering a diode, which I will do.  There is a plate over the terminal pocket, and as I remember from attaching the bell wire, there is plenty of room for a diode.

 

Boy, this sure is fun.  Not only do I get to learn lots of stuff, I get to feel dumb.  I dont use very many FETs.  Just power FETs (IRFZ30/34/44) for stepper motor drivers, and now this project.  I like to use old fashion bi-polar transistors for most stuff because I know how to calculate parameters.  I took a class at the local junior college and about half the semester was about calculating gain and impedance and whatnot for bipolar transistors, so I feel fairly confident using them.  2n2222 or 2n3904/6 being what I have in my parts bin, along with a bunch of high gain transistors I got from my mentor that I never use.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The 2N7002 will work "OK" , since you are driving about 130 ohms relay...might loose about 0.2V across the fet.

 

Take a look at the DMG102T   https://www.diodes.com/assets/Datasheets/ds31783.pdf     only 10 cents at Digikey. if you buy 100

 

 

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am old fashioned and am using the TO-92 version, which is 2N7000.  The 02 is surface mount.  I mostly do perfboard stuff and like the through hole version of things.

 

 

Edit: typo

Last Edited: Tue. Mar 17, 2020 - 06:18 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I do like the gate threshold on that FET has a maximum of 1V.  Most of the FETs I have have much larger gate thresholds.  The IRFZ34 has a max of 4V.  I know I have had trouble in the past getting them to turn on all the way unless I use 5V on the gate.

 

The mayor of San Francisco has put the city on lockdown.  It is now a crime to leave the house except for necessary reasons.  What a load of crap.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarkThomas wrote:
On the one we are using the terminals in the pocket have little tabs on them with holes to solder to. 

Post a picture.  In 30+ years working with mag locks all teh ones I have used have terminal headers with screw terminals for each wire.

 

JIm

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok. I'll try. I am walking down there later today to solder in the Pi protection diode. And be an actor in a video showing how it works to send to the post office. Hopefully i wont get arrested for leaving my house.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


Hi Jim,

Here is a picture.  It is not very good because the lighting was bad and I was probably shaking a little with the long exposure time needed.  You can see the short little tabs under the screws.  There are little holes in the ends of the tabs.  The red wire looks like it attached to the tab, but it goes under it.  You could either solder to the tab, or use it as a blade connector and slide a mating female over it.  Both of the Chinese locks had the little tabs under the screws.  I'm really sorry the picture is so blurry, but it was the best I could do in the time I had.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 It is now a crime to leave the house except for necessary reasons.

You can say you needed to test your lock circuitry laugh

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yeah, that is definitely a serious must do activity.  It turns out the cops are only interacting if they see a group of 6 or more people together, and they order them to disband and separate.  The city is really quiet to walk around in, and there are lots of parking places, which is very unusual.

 

I learned some good stuff from this project I didn't expect to learn.  I might spend a few bucks for some new TO-92 FETS with smaller resistance and gate thresholds, especially if I start getting into 1.8V detectors.  Thank you all for your support and advice, I appreciate it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

the cops are only interacting if they see a group of 6 or more people together,

That's what I call equal opportunity,  non white people have had that privilege for many years. cheeky

 

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I feel honored to be included.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarkThomas wrote:
It turns out the cops are only interacting if they see a group of 6 or more people together, and they order them to disband and separate. 

I'd like to see them try that in this area....We'd laugh at them.  I'd tell them that we are impersonating Rice Krispies in a bowl of milk. (Google George Carlin to understand that one).

 

Anyway.  There is no provision for your soldering those terminals.  oyu are supposed to loosen teh two screws and either wrap you wires under them and tighten them down or use a small fork connector under each screw.  You crimp your power and Diode/mov wires in each connector accordingly.

 

I for one prefer the MOV over the diode because it's not going to short out on reverse polarity...and a coil wont care anyway.

 

Cheerio!

 

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why can i not solder to those tabs? They look designed to be soldered to

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

this is interesting....a lone diode can slow the opening enough to enhance welding (bad).

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=13C3264_AppNote&DocType=CS&DocLang=EN

 

another goody, with some numbers.  They like movs &  also diode-zener combos & TVS devices...but not plain diodes.

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=13C3311_AppNote&DocType=CS&DocLang=EN

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarkThomas wrote:
Why can i not solder to those tabs? They look designed to be soldered to

Because the screws are a giant heat sink. I should have said you shouldn't solder to them. Go for it.

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
a lone diode can slow the opening enough to enhance welding (bad).

That sounds like a bigger problem with high power relays.  I wonder if my little 3V relay would have the same issues.  Interesting stuff.

 

jgmdesign wrote:

Because the screws are a giant heat sink. I should have said you shouldn't solder to them. Go for it.

I could take the tabs off the screws to solder to them to avoid cold solder joints.  With lead solder I still think I could get the tab hot enough to solder to, but you are probably right.

 

Question:  Can I use a standard 1N4148 across the relay coil for protection, or should I be using something like a 1N4007 because of the potentially high voltage?  I have lots of both.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jim,

The problem we had just attaching the bell wire to the door lock was when we unscrewed the terminal screw far enough to get the single wire around it a nut on the inside fell off and it was a pain in the butt getting it back into place to get the screw started.  The screws are too short, in my opinion.  That's why those tabs looked so inviting to solder to.

mark

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0



 That's why those tabs looked so inviting to solder to.

They ARE called solder tabs for a reason! 

At 20ma , any diode will work...it just has to be able to handle the coil current, when it "takes over" the flow

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarkThomas wrote:

Question:  Can I use a standard 1N4148 across the relay coil for protection, or should I be using something like a 1N4007 because of the potentially high voltage?  I have lots of both.

 

I'd use the '4007.  Probably can handle more current, too.  S.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarkThomas wrote:

Jim,

The problem we had just attaching the bell wire to the door lock was when we unscrewed the terminal screw far enough to get the single wire around it a nut on the inside fell off and it was a pain in the butt getting it back into place to get the screw started.  The screws are too short, in my opinion.  That's why those tabs looked so inviting to solder to.

mark

 

I have only ever had to loosen teh screw a few turns then push/pull teh wire underneath the screw head as the screw pops up when you do this.  I do concede it's a pain in the ass and they never changed the design in all these decades but we all do it this way.

 

avrcandies wrote:
They ARE called solder tabs for a reason! 

Yes, and if you look they are under that screw that Mark said is a pain to re-seat to the nut.  That screw is a giant heatsink.  Teh wires are soldered  on during manufacturing and then placed under teh screw head.  Use a big enough iron myabe, at teh cost of melting the case.  But hey ho, go for it.

 

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

Pages