Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Diode across a relay, 1N4148 or 1N4001?

Status
Not open for further replies.
what current will the main coil take?
 
William At MyBlueRoom said:
I've seen both used in schematics. On a small signal < 12V relay coil would a cheap 1N4148 work fine for back EMF?

Depending on a lot of different variables, the back EMF from this type of relay can briefly reach several hundred volts. The 1N4148 has about a 75-volt rating, the 1N4001 is 100 volts. Stick with a 1N4004 to 1N4006 (or equivalent).
 
GizmoWizard said:
Depending on a lot of different variables, the back EMF from this type of relay can briefly reach several hundred volts. The 1N4148 has about a 75-volt rating, the 1N4001 is 100 volts. Stick with a 1N4004 to 1N4006 (or equivalent).

The back EMF from turning an inductor OFF can indeed reach considerable levels, but NOT when the diode is placed across the coil. The didoe conducts on the back EMF, limiting it to only 0.7V, and the dampening effect stops the coil ringing further so voltage isn't much of a problem.

So high voltage diodes aren't needed, and in my experience high current ones aren't usually either?, I've used 1N4148's across relay coils, and I've used mains rectifiers (basically depending what I happen to find first), and I've never seen a failure.

If I was switching a LARGE relay I'd tend to use a rectifer diode, but for normal small relays the 1N4148 works fine.
 
GizmoWizard said:
William At MyBlueRoom said:
I've seen both used in schematics. On a small signal < 12V relay coil would a cheap 1N4148 work fine for back EMF?

Depending on a lot of different variables, the back EMF from this type of relay can briefly reach several hundred volts. The 1N4148 has about a 75-volt rating, the 1N4001 is 100 volts. Stick with a 1N4004 to 1N4006 (or equivalent).

This not correct. With a diode across the coil, the back EMF will be about 0.7 Volt, ie. the forward voltage drop of a diode.

At the moment the coil circuit is opened, the current that was flowing through the coil now goes through the diode. So the diode needs to be able to withstand that current. For example, if the coil current is 20 mA, then a 1N4148 would be adequate. But if it were 900 mA, then a 1N4001 would be required.
 
Further to the above. There are several ways to limit the back EMF.

The disadvantage of the diode solution is that it increases the relay release time. Often this is not critical, but if it is the following options are available:-

1. A resistor across the coil in lieu of the diode. This reduces the time constant and the back EMF is R * I where R is the resistor value and I is the coil current. For example, a 12 Volt relay with a 500 Ohm coil will draw about 24 mA. So if a 1k resistor is connected across the coil, the peak EMF will be 24 mA * 1k = 24 Volt. The time constant will be reduced from L/500 to L/1500.
Where L is the coil inductance.

The disadvantage of this solution is that there is current through the resistor while the relay is operated and thus energy is being consumed unnecessarily. This can be avoided by inserting a diode is series with the resistor. Then there is no current through the resistor while the relay is operated. But the diode is switched on when the circuit is opened and the EMF will be about 24.7 Volt using the figures from the above example.

2. Connect a diode in series with a Zener diode across the coil. This limits the back EMF to about Vz + 0.7 Volt. In this case, most of the energy stored in the magnetic field is dissipated in the Zener. So the peak power dissipation in the Zener needs to be considered.

3. connect a resistor and capacitor in series across the coil. The C & R values need to be calculated so that the circuit is critically damped.

If it is over damped, the release time is increased, if under damped, there will be a damped oscillation which may not be desirable.
 
nigel you still don't get it if the diode has a low saturation voltage {reverse} to you it will breakdown and go south to mexico for a vacation and the voltage will not be .7v or whatever but a short. god you guys still learning i supose. imagine a 75 volt spike and 20ma figure it out the power and you will get the an answer fast. learn my friend.
 
neon said:
nigel you still don't get it if the diode has a low saturation voltage {reverse} to you it will breakdown and go south to mexico for a vacation and the voltage will not be .7v or whatever but a short. god you guys still learning i supose. imagine a 75 volt spike and 20ma figure it out the power and you will get the an answer fast. learn my friend.

I don't know what your problem is?, but you appear to be attacking me for no reason?, and not only that with no justification - you are wrong about this type of circuit. The back EMF only appears as the field collapses, so you NEVER get 75V across the diode, in any direction.

1N4148's have been used for decades across relay coils, and never been a problem, as a little theoretical knowledge easily shows.
 
neon said:
nigel you still don't get it if the diode has a low saturation voltage {reverse} to you it will breakdown and go south to mexico for a vacation and the voltage will not be .7v or whatever but a short. god you guys still learning i supose. imagine a 75 volt spike and 20ma figure it out the power and you will get the an answer fast. learn my friend.
I'm afraid it is you that does not understand.

As I said above, the back EMF is limited to about 0.7 Volt.

When the circuit is opened, the inductance wants to keep the current flowing in the same direction, so the diode is FORWARD biassed, not reverse biassed.

I suggest that you study Lenz's Law in your reference books. It states:- Induced EMFs and their resultant currents are in such a direction as to oppose the motion that produced them.

In this case, the "motion" is the decaying magnetic field in the relay's magnetic circuit.

The back EMF is therefore in such a direction as to keep the current flowing in the same direction as it was when the transistor was on. Thus the collector end of the coil becomes positive and so the diode conducts in its forward direction.

If you observed the collector voltage on a storage scope, it would rise from the saturation level to about 12.7 Volt.

You can confirm this by doing it in practice or by using a software simulation package such as Switcher CADIII.

I spent many years of my life designing complex electromechanical systems and interfacing electronics, hence the comments in my previous post.

So please don't make smug comments as they are not called for when you are right and can come back to bite you when you are wrong - none of us is the fountain of all knowledge, we all make mistakes occasionally..
 
I know this an old thread, but I'll give this a shot any way.

Do you have to be concerned with the back emf on an AC solenoid, as opposed to a DC relay, etc?

If so, I suppose the resistor would help, but it would cause additional current draw.

The coil in question is a 24vac sprinkler valve, which draws 190 ma while holding the valve open.
 
What is switching the valve, a triac? If so, it turns off at a zero-crossing of the AC input, so usually no suppression is needed.
 
Do you have to be concerned with the back emf on an AC solenoid
Yes. Back-to-back zeners across the coil will suppress the spikes.
 
Yes, in series.
What is switching the valve, a triac? If so, it turns off at a zero-crossing of the AC input, so usually no suppression is needed.
Usually un-needed, but it would depend on the coil inductance since the coil current is out of phase with the AC input and may be relatively high when the AC voltage crosses zero.
 
Last edited:
I have taken apart a sprinkler valve timer which uses triacs and it had no suppression. The inductance of the solenoid coil likely is quite small. It is wound with tiny wire, so mostly the coil resistance limits the coil current.
 
What is switching the valve, a triac? If so, it turns off at a zero-crossing of the AC input, so usually no suppression is needed.


Sorry so long to respond. I should have cleared this site with my spam box.

Each valve is being switched with a set of n.o. relay contacts. Those intermediate relays are DC and each protected by a small diode across each coil So there is no synchronization with the sine wave. Perhaps the synchronization issue is a good reason to use triacs instead. But a little 8 relay board was $8 and it is complete with opto isolators.

Thank you
 
Yes. Back-to-back zeners across the coil will suppress the spikes.

So perhaps 40 volt zeners placed in series. So in each direction, one conducts with a theoretical maximum voltage of say 40.7 v across the coil? I hadn't thought of doing something like that.

Thanks.
 
I have taken apart a sprinkler valve timer which uses triacs and it had no suppression. The inductance of the solenoid coil likely is quite small. It is wound with tiny wire, so mostly the coil resistance limits the coil current.

That is good news! I am just trying to keep from frying the little web server board and also the contacts of the controlling relay. And, of course, I would prefer to minimize the component count.

I am currently looking at a four-channel interface module out of a Hunter Sprinkler Controller. It seems a small control voltage would be sent to each of the four channels of the module to turn each on. Further, it appears there is a small coil, a varistor, a triac, a small transistor, and perhaps 3 resistors for each valve connection. It appears one side of each varistor is grounded. Neither side of the coils is grounded. The parts are really crammed together.

Thank you for the advice.
 
I just went and measured one of these. The DC resistance is 33Ω. The inductance is ~30mH.

I dont think that you need any suppression. The current being interrupted is <1A, which should be well within the AC inductive load rating of most small relays.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top