Need a Diode for SPDT Relay With Coil Operated Only by Switch?

Thread Starter

Thenextman

Joined Feb 3, 2017
42
Hello all,

I have built a circuit which uses this automotive relay: https://www.amazon.ca/MICTUNING-5-Prong-Harness-Socket-Color-labeled/dp/B01MDLX3BW/ref=sr_1_1?s=automotive&ie=UTF8&qid=1495023392&sr=1-1&keywords=relay along with a momentary switch (to activate the coil) and a 12VDC 1 A wall wart.

Everything worked great for a while, but the client is having mysterious issues after about 40 hours of use. I cannot troubleshoot effectively unfortunately because the unit is currently in storage.

I have attached an image of my original circuit - some things to keep in mind:
  1. I do not know how to tell + from - on the PSU, and do not know the standard for black and white on the relay harness, and do not know which way the diode must be pointing if it is required. I do have a voltmeter. if you think the lack of diode caused the issue, I will be rebuilding the circuit, so any indication on how to properly do this would be phenomenal.

  2. The PSU is connected only to the momentary switch and the relay coil - no power is being applied across the switching contacts of the relay (30 / 87 / 87a).

  3. My hope is that wired like this, the diode isn't really necessary, but will redo this if needed of course.
Thanks all - looking forward to your thoughts!
 

Attachments

drc_567

Joined Dec 29, 2008
688
It looks like you need to reverse the diode polarity, so that the negative ... bar ... is facing the positive + terminal. This will prevent damage to the coil, which may have occurred if the switch was used frequently. Check the relay coil for continuity, and that there is no ground to the frame.
 

Thread Starter

Thenextman

Joined Feb 3, 2017
42
It looks like you need to reverse the diode polarity, so that the negative ... bar ... is facing the positive + terminal. This will prevent damage to the coil, which may have occurred if the switch was used frequently. Check the relay coil for continuity, and that there is no ground to the frame.
Thanks drc. In your mind, the diode is absolutely necessary?

When using my voltmeter, how can I tell what is +/- on the PSU? Is "neutral" + or - in a DC circuit? Do white and black mean +/-?

Thanks again
 

EM Fields

Joined Jun 8, 2016
583
Hello all,

I have built a circuit which uses this automotive relay: https://www.amazon.ca/MICTUNING-5-Prong-Harness-Socket-Color-labeled/dp/B01MDLX3BW/ref=sr_1_1?s=automotive&ie=UTF8&qid=1495023392&sr=1-1&keywords=relay along with a momentary switch (to activate the coil) and a 12VDC 1 A wall wart.

Everything worked great for a while, but the client is having mysterious issues after about 40 hours of use. I cannot troubleshoot effectively unfortunately because the unit is currently in storage.

I have attached an image of my original circuit - some things to keep in mind:
  1. I do not know how to tell + from - on the PSU, and do not know the standard for black and white on the relay harness, and do not know which way the diode must be pointing if it is required. I do have a voltmeter. if you think the lack of diode caused the issue, I will be rebuilding the circuit, so any indication on how to properly do this would be phenomenal.

  2. The PSU is connected only to the momentary switch and the relay coil - no power is being applied across the switching contacts of the relay (30 / 87 / 87a).

  3. My hope is that wired like this, the diode isn't really necessary, but will redo this if needed of course.
Thanks all - looking forward to your thoughts!
1. the connections to the coil aren't polarity sensitive. What matters is that you get 12 volts across the
coil when you want to energize the relay. The diode should be connected so that its cathode is connected to +12 volts and its anode to 0 volts. The way to tell what's what is to connect your voltmeter across the PSU and note whether the reading is +12 or - 12. If it reads +12 then the voltmeter's +lead is connected to the PSU's + terminal. If it reads -12, then the voltmeter's - lead is connected to PSU +12.
Color codes mean nothing unless you have a data sheet on hand.

2. OK

3. A diode isn't usually needed if the relay coil is being switched mechanically instead of with a transistor.
Automotive relays made by reputable manufacturers are pretty rugged beasts when dealing with the contact voltages and currents they're designed for, so it may very well be that the way they're being used by your client is bogus. It also may be that the switch you're using is a very low-current affair and is failing.

If you want definitive answers instead of just guesswork, you need to post data sheets for the relay and the switch and post a schematic showing how the relay's being used.

4. What are the "mysterious issues"?
 
Last edited:

drc_567

Joined Dec 29, 2008
688
... kind of depends on the amount of current that is being switched. ... Better, longer life usually with one. The diode allows the current in the coil to 'recirculate' as the switch is opened.
Without the diode, the current flow comes to an abrupt halt ... causing an undesirable, sometimes damaging, voltage.

What is the diode that you are using? They come in different current sizes.

With your voltmeter, set to volts, and see which probe arrangement shows '+' voltage. The red probe, if correctly placed, should be '+'. The black probe, inserted in the 'com' terminal will be '-'.

Don't rely too much on the circuit wire colors. Sometimes they are ok, sometimes not.
 

Thread Starter

Thenextman

Joined Feb 3, 2017
42
1. the connections to the coil aren't polarity sensitive. What matters is that you get 12 volts across the
coil when you want to energize the relay. The diode should be connected so that its cathode is connected +12 volts and its anode to 0 volts. The way to tell what's what is to connect your voltmeter across the PSU and note whether the reading is +12 or - 12. If it reads +12 then the voltmeter's +lead is connected to the PSU's + terminal. If it reads -12, then the voltmeter's - lead is connected to PSU +12.
Color codes mean nothing unless you have a data sheet on hand.

2. OK

3. A diode isn't usually needed if the relay coil is being switched mechanically instead of with a transistor.
Automotive relays made by reputable manufacturers are pretty rugged beasts when dealing with the contact voltages and currents they're designed for, so it may very well be that the way they're being used by your client is bogus. It also may be that the switch you're using is a very low-current affair and is failing.

If you want definitive answers instead of just guesswork, you need to post data sheets for the relay and the switch and post a schematic showing how the relay's being used.

4. What are the "mysterious issues"?
Thanks EM and drc

The switch is this unit http://www.tapeswitch.com/Mats/CVP.html

The power supply is this https://www.amazon.ca/gp/aw/d/B01LXJE7HR/ref=ya_st_dp_summary

No data sheet for the relay, but there are some specs if you open the description here https://www.amazon.ca/gp/aw/d/B01MDLX3BW/ref=cm_cr_arp_mb_bdcrb_top?ie=UTF8#productDescription_secondary_view_div_1495028949199

Basically when someone stands on the switch mat, it close the loop between 2 and G on the other component. Otherwise, 1 and G are a closed loop.

The client has been having issues where it seems that the switch times out - I.e. It switches over but then switches back without the person releasing the switch.

If 1 and 2 are simultaneously connected to G on the other component, this may cause issues. Not sure if that's possible unless the SPDT is arcing or something?

Given the components above, do you think I need to send them a new relay with a flyback connected? Is it possible the switch is fried? This would be much more difficult to replace.

Thanks again!
 

shortbus

Joined Sep 30, 2009
6,885
Why not use an automotive relay with a built in diode? Some are made that way. Then use another diode in the input line to that one, this would then only allow the circuit to work when it is properly hooked to the power supply. An image of the circuit of a relay with built in diode.
upload_2017-5-17_10-43-16.jpeg
 

Raygl

Joined Dec 3, 2016
10
Is it possible that not using the diode or some type of snubber that the arc from the coil is destroying the switch?
Even for a switch I would use the diode. If the high voltage side is inductive, I would recommend some type of snubbing circuit.
 

Tonyr1084

Joined Sep 24, 2015
3,470
Was this not made clear in your last post? The diode goes against the positive. The bar on the end of the diode faces the positive to block DC from flowing to ground. When power is interrupted the coil of the relay generates a flyback voltage that can damage sensitive electronic components. Since you're using a tape switch (floor mat) I'd recommend the diode. But today is Thursday and you were saying you needed to have this solution in your customers hands by today. What's going on here? Once again, here is a drawing showing how to place the diode:

Flyback Diode Coil.png
 

Tonyr1084

Joined Sep 24, 2015
3,470
Diodes flow current in one direction (with few exceptions). When you want a positive flow of current from an alternating source you use a diode to block the reverse current direction. The result is a pulsating direct current. You can use a full wave rectifier (four diodes [not shown]) to turn the negative current into a positive direct current. However, it will still be subject to peak voltages and zero voltages (not fully explained here). Adding a capacitor can smooth out those pulses and result in a DC ripple current. All using diodes and a capacitor.

In the case of a flyback (or snubber) circuit you want to block DC from going to ground through the diode. Hence, it's negative (bar end) faces the positive source. When current is interrupted the magnetic field in the coil breaks down and produces a current of its own, usually in a much higher voltage. The diode will act as a short circuit and "Snub" the current and prevent it from going anywhere else. Hence, the flyback voltage is snubbed by the snubber diode.

Let me draw you a picture of a diode and what the markings mean:

DIODES.png
 

EM Fields

Joined Jun 8, 2016
583
Diodes flow current in one direction (with few exceptions). When you want a positive flow of current from an alternating source you use a diode to block the reverse current direction. The result is a pulsating direct current. You can use a full wave rectifier (four diodes [not shown]) to turn the negative current into a positive direct current. However, it will still be subject to peak voltages and zero voltages (not fully explained here). Adding a capacitor can smooth out those pulses and result in a DC ripple current. All using diodes and a capacitor.

In the case of a flyback (or snubber) circuit you want to block DC from going to ground through the diode. Hence, it's negative (bar end) faces the positive source. When current is interrupted the magnetic field in the coil breaks down and produces a current of its own, usually in a much higher voltage. The diode will act as a short circuit and "Snub" the current and prevent it from going anywhere else. Hence, the flyback voltage is snubbed by the snubber diode.

Let me draw you a picture of a diode and what the markings mean:

View attachment 127067
Just for grins, here's a picture of what happens - with and without a catch diode - when a relay coil (or any inductance) is energized, current allowed to reach steady state through the inductor and its intrinsic resistance, and then the circuit quickly opened. :
Relay coil damping with catch diode.png Yabbut MEGAvolts????
Sure, since since E = L di/dt, if dt gets small enough, E will tend toward infinity.
 

Thread Starter

Thenextman

Joined Feb 3, 2017
42
Why not use an automotive relay with a built in diode? Some are made that way. Then use another diode in the input line to that one, this would then only allow the circuit to work when it is properly hooked to the power supply. An image of the circuit of a relay with built in diode.
View attachment 127018
Wish I had known this in the first place! The issue now is that the
Diodes flow current in one direction (with few exceptions). When you want a positive flow of current from an alternating source you use a diode to block the reverse current direction. The result is a pulsating direct current. You can use a full wave rectifier (four diodes [not shown]) to turn the negative current into a positive direct current. However, it will still be subject to peak voltages and zero voltages (not fully explained here). Adding a capacitor can smooth out those pulses and result in a DC ripple current. All using diodes and a capacitor.

In the case of a flyback (or snubber) circuit you want to block DC from going to ground through the diode. Hence, it's negative (bar end) faces the positive source. When current is interrupted the magnetic field in the coil breaks down and produces a current of its own, usually in a much higher voltage. The diode will act as a short circuit and "Snub" the current and prevent it from going anywhere else. Hence, the flyback voltage is snubbed by the snubber diode.

Let me draw you a picture of a diode and what the markings mean:

View attachment 127067
You guys are going to be proud of me.

Picked up this relay (MZ-12HS-U) http://www.mouser.com/ds/2/164/mz-469330.pdf

Diode is 1N4007.

Will be hooking up as in the attached photos.

Let me know what you think! Will wire up tomorrow or Saturday depending on your feedback.

Is it ok to insert the diode so scrunchy like?

EM. I saw a video on the voltage spikes with no diode. Scary. Hopefully nothing has arc welded in the switch. The symptoms would say otherwise (would expect it to get stuck on input 2) but you never know how accurate client feedback is.

Thanks guys!image.jpgimage.jpg
 

Attachments

Last edited:

EM Fields

Joined Jun 8, 2016
583
PPS just tested the psu. With the red lead touching the dashed line and com touching the wire with writing on it, negative reading. When the red lead touches the wire with the writing and the black lead (com) touches the wire with the dashes, positive reading. So. I guess I assumed wrong!? Wire with writing is +?
Yes.
 

Tonyr1084

Joined Sep 24, 2015
3,470
Great! Just one VERY IMPORTANT question on your drawing: Which is the positive lead coming from the power supply?

Get it wrong and you will lose either the power supply or the diode. OR both.

And one not so important question: I thought you had to resolve this by - um - yesterday?
 

Tonyr1084

Joined Sep 24, 2015
3,470
Want first hand experience with flyback? Take a small transformer and a double A battery. Using your fingers, touch the wires from either the primary or the secondary to the battery. Then disconnect from the battery.

WARNING! YOU WILL GET A SHOCK! Yes, from a double A battery and a coil.

Then try it again but with a flyback diode in place. No shock. None whatsoever that you can feel.
 

EM Fields

Joined Jun 8, 2016
583
Thanks EM. What do you think about that circuit?
Looks good.
You should edit your schematic though, since you found that the PSU wire with the dashed line is the supply return, which means the other one should be connected to the diode's cathode through the switch. Like this, maybe:
Mat switch circuit.png
 
Last edited:

Thread Starter

Thenextman

Joined Feb 3, 2017
42
Great! Just one VERY IMPORTANT question on your drawing: Which is the positive lead coming from the power supply?

Get it wrong and you will lose either the power supply or the diode. OR both.

And one not so important question: I thought you had to resolve this by - um - yesterday?
So the PSU has two cables coming out, one with a dashed line on it, and one with writing on it. I took my voltmeter, and touched the red lead to the cable with writing, and the black (com) lead to the cable that is dashed, and I get a positive voltage. I am assuming that means cable with writing is positive, so I will wire as drawn in attached. Look right?

Next Thursday must be with client.

So crazy that you can shock yourself with low voltage!!

Thanks Tony
Looks good.
You should edit your schematic though, since you found that the PSU wire with the dashed line is the supply return, which means the other one should be connected to the diode's cathode.
Like this, maybe:
View attachment 127153
Excellent - thanks @EM Fields and @Tonyr1084

I will copy your schematic EM and just assume that placing the switch on the source side is best!

Trust that I've identified source and return / + And - / v+ and g / ... Using the voltmeter. Why is there so many terms for the same thing, I'll never understand!

Thanks again
 
COMMENTS: (In the hard details).

For automotive relays there is a SUGGESTED positive coil circuit number. This would be so the relays with internal diodes can be substituted. (No one pays attention to this).

Automotive relays may have the same numbers, but are wired differently. 200 PIV for the diodes is a decent value to use. Prewired sockets (single and dual relays) with wire leads are available.

A small ceramic cap across the switch might help close to the switch.

You can likley get away with soldering the diode across the coil pins close to the body of the relay.

Here http://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=V23134-X0000-A001&DocType=DS&DocLang=English you can see various versions of an automotive relay

Relays have various specs. Note that this one has a "minimum contact load of" 5V @ 1A. Flakey results could happen if this is not respected. To get the minimum loads lower, usually more exotic materials are used for the contacts.

If your not given the "wetting current: specs, 10 mA ia a good number to assume, but this one is 1 A @ 5V.
 
Top