Im using these valves 12VDC solenoids
http://www.dudadiesel.com/choose_item.php?id=W160B15

Im using a relay to control the power to them. Should I put in protection diodes on the solenoids to prevent any high voltage on power off a solenoid? I dont want another things using the same 12VDC source to be hit with anything bad.

 

If there are any semiconductors on the same power line then YES.

transistors, microprocessors, (TV's, CNC machines, computers, etc.) all need protection from reverse voltage spikes.

If the solenoids are the only thing on the power circuit then you can do without them. But the poor guy who comes along in a few weeks or years and tries to use that power circuit will get a surprise

 

For small solenoids, motors etc. a 1N4148 usually suffices. For other types, it becomes much more difficult...

 

I have a used a solenoid for a simple electric door lock. It works on 12V battery. The solenoid is connected to a push-button switch in series with the battery. There is a string that pushes the metal core of the solenoid out, when the switch is pressed, the core is attracted to inside... Now the problem that I'm facing with that is that the solenoid seems to be charged or magnetized and after the core is attracted when the switch is pressed, it stays inside even after I release the push button switch, where the string is forcing the core outside but it seems to be slightly charged/magnetized and attracted to inside. A very small hit makes the core goes outside. I'm sure it's not a mechanical problem (friction, string...).

I did put a diode in reverse polarity of the battery on the terminals of the solenoid, but it has no effect on the problem.. Any idea?

Regards
Hazim

 

I believe the correct way to ask a question is to start your own topic and post the question there and not place it inside another persons topic.

Try doing that and include in the question what materials you have used to make the solenoid and plunger from. If the plunger is retaining magnetism then it is not made of the proper material.

 

Common answer:

Use something about 4 x the coil voltage and 10x the coil current - minimum.

The door latch may have become magnetized but at first it didn't have one of these diodes and the plunger has become highly magnetized.

 

I have some ICs on the same 12v rail. Does it make sense to put one big diode between 12V and GND? These things run @12V x 2.6 AMPs and I have 10 od them. So I figure @ 10x the current I would need a 48V 260AMP diode. Does that make sense? does one even that big exist? Not sure what to do.

 

Looks like maybe an IN3737 would work? Seems expensive BUT if this is the best way... I could just buy 10 SEMIKRON P2500G (25AMPs) and put them on each solenoid... Suggestions?

 

It's best to put one diode at each of the sources, you want the current path to be as short as possible.

And you don't figure out the value by what the solenoid or motor draws, what you're dealing with is the reverse EMF which will usually be much higher in voltage and current than the original activating source. The peak voltage and current is what you're trying to stop.

As a rule I try to use something that's at least 4 x the driving voltage and has well over 10 x the surge current response so a common/inexpensive 1N4004 on each should be fine. Actually about any rectifier normally rated for 1A at 100V will have sufficient surge current ratings so long as the circuit isn't being switched at a high repetitive rate. Larger never hurt though, the 1N5400 series is a handy and rather durable rectifier to keep around in your parts bin.

 

A diode between 12V and GND will not help.

For such high current coils you should really consider PWMing them.
http://focus.ti.com/lit/ds/symlink/drv101.pdf
http://www.st.com/stonline/products/literature/ds/1332.pdf

 

Didn't even notice that was mentioned. The protection diodes must be across the device(s) that generate the emf - in this case the solenoids themselves.

 

The flyback voltage from a switched inductor does not appear on the supply rail. It appears on the switched side of the inductor.
Also, the protection diode peak current will not exceed the magnitude of the current that was flowing through the inductor when it was switched off.

 

Distributor ignition systems and fuel injector circuits must operate quickly. They do not have back emf diodes across their coils. They have zener diodes across their drivers.

 

the protection diode peak current will not exceed the magnitude of the current that was flowing through the inductor when it was switched off.

Physics tells us that but I've always stuck to my 4x / 10x rule to allow plenty of safety margin. Peak handling current of a rectifier rapidly decreases with every pulse that heats up the junction, this was especially true with the older rectifiers.

 

I haven't really checked it out in any detail but larger diodes generally turn on much slower giving you a potentially larger spike.

 

I haven't really checked it out in any detail but larger diodes generally turn on much slower giving you a potentially larger spike.

Diode turn on delay is never a problem in this type of application (solenoid or relay) even with slow rectifier diodes.

See: http://www.cliftonlaboratories.com/diode_turn-on_time.htm

 

Diode turn on delay is never a problem in this type of application (solenoid or relay) even with slow rectifier diodes.

See: http://www.cliftonlaboratories.com/diode_turn-on_time.htm

Good to know, thanks. Posts like this justify the amount of time I spend here lately

 

Great info. I put a decent size diode on each solenoid and I dont have spikes anymore. Without them my 12V line was going over 20V! Question though. Why PWM them?

 

...Question though. Why PWM them?

If you ever want the solenoids to be partially on, then this may be a good idea, but since you said that you are currently using a relay to turn them on/off, this must have been an answer to some other question.

Relatively substantial diodes, directly across the solenoids, as near to the physical location of the solenoids as is reasonable, is what you want.

 

If you ever want the solenoids to be partially on, then this may be a good idea,...

Solenoids partially on? That's a new one on me.

I expect that whomever mentioned PWM with respect to driving coils was referring to, well, coil drivers (for lack of a better name). Purpose built circuits or IC's which allow a large current for energizing a coil (opening a valve), and then provide a reduced "holding current" for holding an electromagnetic field (holding a solenoid/valve open/closed). PWM is a commonly employed technique for regulating the voltage/current (but certainly not the only way). This type of coil driver is especially necessary with high speed valves ie:1kHz or greater. And generally a good idea if your valve/solenoid costs a lot of money.

In case you didn't know, it takes a large initial current to charge a coil (they act like a short at t=0), but after a magnetic field is created the current necessary to "hold" or keep the field is usually much lower (especially in relays/solenoids).

Here is an example IC: http://www.st.com/stonline/products/literature/bd/14799.htm

PS, Google for "Snubber Diode". You'll find more information about dealing with the Back-EMF caused by discharging coils. And, put your snubber diode as close to the coil as you can. Some relays even come with snubbers built in.