Best Relay

[audiomick]

On 3/15/2023 at 2:02 AM, docc said:

I admit I do not understand from where the voltage spike would be coming from and across what terminals of the relay the diode would block it.

The spike is caused by the same physics that allow a generator, an electric motor, a loudspeaker and your Shure microphone to work.

Wiki article here:

https://en.wikipedia.org/wiki/Magnetic_field#Interactions_with_electric_currents

Very quick summary:

a current in a coil will produce a magnetic field and exert a force on magnets and magnetic objects within the field. This way round is what makes electric motors, loudspeakers and solenoids work.

when a coil (or even a straight bit of wire) is within a magnetic field, a change in the magnetic field will generate an electric current in the coil (or wire). This way round is how generators and dynamic microphones work.

A change in the magnetic field may be caused by moving a magnet past the coil, or the coil in relation to the magnetic field (generators, microphones...), or it may be caused by altering (switching off, for instance...) the current in a coil that is generating the field. The latter is how ignition coils work: the primary coil generates a magnetic field, the points open (ok, the ECU fires the ignition...), current in the primary is turned off. This changes the magnetic field, causing a current to flow in the secondary coil and fire the spark plugs.

In the case of the solenoid, the voltage spike is kind of a "kick-back". The coil is generating the magnetic field to actuate the solenoid. Even though it is generating the field itself, current will still flow in the coil as the magnetic field collapses.

As I understand it, the resulting current flows in the reverse direction to the actuating current, i.e. +Volts (much more than 12 Volt, but very low current) out of contact 85. Looking at the diagrams, the forward direction of the diode is from contact 85 to contact 86. That means, in normal operation the diode is "not conductive" to the +12V that is on contact 86. When + Volts are present at contact 85 (from the spike we are considering) the diode presents effectively no resistance, so the amperes flow down there instead of trying to go through any electronics that might be connected to contact 85.

I'm guessing here, but I assume the voltage peak just shorts itself out through the diode against the other end of the coil.

The rest of the discourse is a combination of "basic electronics for music students" that I learnt at Uni 40 years ago, and stuff out of that e-Hack document that I linked further up. If I got it wrong anywhere, please correct me.

[docc]

That makes perfect sense, and explains why @Speedfrog sees that vulnerability from Relay#5's coil powered by the same feed as the ECU "downstream" from Relay #4.

So some questions, do this spikes only occur when the coil is powered on and off, or do they occur continually?

Also, is the magnitude of the spike amplified by the size of the coil? Always thinking "bigger is better," the OMRON G8HE have 1.8 watt coils, the CIT and PC that are refeneced are 1.5W (the CIT I am using are 1.2W). Perhaps the smallest effective coil is better as long as there is no "bounce" (Loss of contact from vibration or impact)?

Finally, do the resistors serve some of this function by "buffering" the spikes? Enough to be adequately effective?

[audiomick]

8 hours ago, docc said:

So some questions, do this spikes only occur when the coil is powered on and off, or do they occur continually?

only when the coil is switched off

8 hours ago, docc said:

 

Also, is the magnitude of the spike amplified by the size of the coil?

probably, but I suspect that is not something to worry about too much.

8 hours ago, docc said:

 

Finally, do the resistors serve some of this function by "buffering" the spikes? Enough to be adequately effective?

The resistor serves the same function. I spoke to my electronic technician colleague about it earlier. He confirmed that the purpose of the resistor is the same as that of the diode, but that resistors are not quite as effective as the diode in this function.

I didn't follow his explanation very well, though. What I think is, the switching voltage sees a coil with low resistance and a resistor with higher resistance than the coil, so most of the amps go through the coil. The spike, coming out of the other end of the coil, sees the resistor and whatever electronics might be there with a much, much higher input resistance than the resistor is presenting, so most of the amps go through the resistor back to the other end of the coil and "short out".

Don't know for sure if that is right, but it makes sense according to my basis electronic knowledge.

 

[audiomick]

I was guilty of some sloppy language in that last post, volts where I meant amps.

 

It is corrected now.

[Speedfrog]

15 hours ago, docc said:

vulnerability from Relay#5's coil powered by the same feed as the ECU "downstream" from Relay #4.

I'm not sure I understand that description.

On 3/14/2023 at 6:02 PM, docc said:

I admit I do not understand from where the voltage spike would be coming from and across what terminals of the relay the diode would block it.

First let's establish that the Voltage Spike happens at the coil side of the relay (low current control circuit), it does not affect the switch side (high current load circuit).

When the relay is turned off and the coil is de-energized, the magnetic field collapses, resulting in a voltage surge in the opposite direction. These low current surges can have significantly high voltages, often up to 100 volts, with the potential to destroy a transistor driving a relay.

Think of it as a taser, high voltage/low current, it won't kill you but it'll disable you for a while... do it repeatedly or for too long and it'll kill you, same as lightning.

A diode(often called a flyback diode) installed in parallel with the coil (pin 85/86) in reverse polarity creates a low resistance loop for that momentary voltage spike to be absorbed/dissipated and protect the controlling electronic circuit while allowing the current to flow directly through the coil when the relay is switched on.

So you might think, great, let's use diode protected relays whenever we have a transistor on the control side... well, like everything electronic, it ain't that simple.

Although very effective, one drawback to using a flyback diode as a voltage surge suppressor in a relay is that it decays the magnetic field of the coil slower than with no diode, taking longer for the relay to open the contacts and allowing arcing and micro-welds to occur between the contacts which could cause the relay to stick overtime when the contacts become welded together.

I believe the built-in diodes found in micro relays are just simple flyback diodes. There are other types of diodes or diode combinations that can be used to alleviate this issue but it's getting a bit more technical that I'm willing to delve into here.

Resistors installed across the coil provide similar protection against voltage spikes and are more durable if not as effective.

A downside to using a resistor is that it will allow current to flow through it whenever the relay is on and dissipate energy and convert it into heat.

The advantage it holds over a diode is that it won't slow down the opening of the contacts and it is not sensitive to polarity.

It would be interesting to know what type of relays were spec'd by the Guzzi engineers and whether the ECU has built-in voltage surge protection.

The only thing I see on the schematic is a "safety diode" #48 between the ECU relay pin #85 and the ECU, presumably to give it some protection. None of the relays show any suppressor across their coil.

Of course, with our bikes being 20+ years old, I doubt there are many original relays left under the seat.

[docc]

6 hours ago, Speedfrog said:

I'm not sure I understand that description.

First let's establish that the Voltage Spike happens at the coil side of the relay (low current control circuit), it does not affect the switch side (high current load circuit).

When the relay is turned off and the coil is de-energized, the magnetic field collapses, resulting in a voltage surge in the opposite direction. These low current surges can have significantly high voltages, often up to 100 volts, with the potential to destroy a transistor driving a relay.

Think of it as a taser, high voltage/low current, it won't kill you but it'll disable you for a while... do it repeatedly or for too long and it'll kill you, same as lightning.

A diode(often called a flyback diode) installed in parallel with the coil (pin 85/86) in reverse polarity creates a low resistance loop for that momentary voltage spike to be absorbed/dissipated and protect the controlling electronic circuit while allowing the current to flow directly through the coil when the relay is switched on.

So you might think, great, let's use diode protected relays whenever we have a transistor on the control side... well, like everything electronic, it ain't that simple.

Although very effective, one drawback to using a flyback diode as a voltage surge suppressor in a relay is that it decays the magnetic field of the coil slower than with no diode, taking longer for the relay to open the contacts and allowing arcing and micro-welds to occur between the contacts which could cause the relay to stick overtime when the contacts become welded together.

I believe the built-in diodes found in micro relays are just simple flyback diodes. There are other types of diodes or diode combinations that can be used to alleviate this issue but it's getting a bit more technical that I'm willing to delve into here.

Resistors installed across the coil provide similar protection against voltage spikes and are more durable if not as effective.

A downside to using a resistor is that it will allow current to flow through it whenever the relay is on and dissipate energy and convert it into heat.

The advantage it holds over a diode is that it won't slow down the opening of the contacts and it is not sensitive to polarity.

It would be interesting to know what type of relays were spec'd by the Guzzi engineers and whether the ECU has built-in voltage surge protection.

The only thing I see on the schematic is a "safety diode" #48 between the ECU relay pin #85 and the ECU, presumably to give it some protection. None of the relays show any suppressor across their coil.

Of course, with our bikes being 20+ years old, I doubt there are many original relays left under the seat.

My original relays are in the their prescribed location: shelved in the pile cabinet, properly labeled , one example of seven generations of relays I have run on mySport (I skipped the TYCO, not haven fallen for that).

[audiomick]

21 minutes ago, docc said:

My original relays are in the their prescribed location: shelved in the pile cabinet, properly labeled...

Would you care to measure and report the physical dimensions of the original relay, just for interest's sake?

 

If you can manage inches and millimetres, that would be brilliant...

[docc]

32 minutes ago, audiomick said:

Would you care to measure and report the physical dimensions of the original relay, just for interest's sake?

 

If you can manage inches and millimetres, that would be brilliant...

When the new Picker Components relays arrive (in a few days from onlinecomponents), I am planning a V11 RelayFest /TechSession.  I can tell you the original Seimens are small, and the very first warranty/dealer support I received was a larger Seimens to fix in the Starter Relay/#1 position, as the troubles were already being reported.

I'll report back with dimensions, yet we certainly found weight is an indicator:

" . . . [OMRON G8HE] 20.0 grams with my analog method. For comparison, my original Seimans are 15.1 grams."

 

[audiomick]

On 3/14/2023 at 7:22 PM, audiomick said:

Thanks.

In the meantime I had a look at this again:

https://www.e-hack.de/der-grosse-kfz-relais-ratgeber/

I just found a document quite similar to that, but in English.

https://www.electronicshub.org/relay-wiring-diagram/

[Speedfrog]

1 hour ago, audiomick said:

I just found a document quite similar to that, but in English.

https://www.electronicshub.org/relay-wiring-diagram/

Good primer on relays, but no mention of suppressor, diode or resistor, or flyback voltage spikes.

[Speedfrog]

5 hours ago, docc said:

My original relays are in the their prescribed location: shelved in the pile cabinet, properly labeled , one example of seven generations of relays I have run on mySport (I skipped the TYCO, not haven fallen for that).

I would be interested to know if the original relays were all the same in all positions and if there is any marking on them indicating the presence of a surge suppressor.

[audiomick]

15 minutes ago, Speedfrog said:

Good primer on relays, but no mention of suppressor, diode or resistor, or flyback voltage spikes.

Yes, it is for beginners. Like me.

9 minutes ago, Speedfrog said:

I would be interested to know if the original relays were all the same in all positions and if there is any marking on them indicating the presence of a surge suppressor.

Me too.

[Speedfrog]

@audiomick Here is a good read if you are interested to go beyond basics.

https://www.12voltplanet.co.uk/relay-guide.html

[docc]

5 hours ago, Speedfrog said:

I would be interested to know if the original relays were all the same in all positions and if there is any marking on them indicating the presence of a surge suppressor.

I'm not sure I understand the value of investigating the original relay as it was proved failure prone from mid-2000 year.

Yet, my 2000 Sport was delivered with the same relay in all four 4-pin (SPST) positions, and a different 5-pin (SPDT) in position #1. As mentioned, my dealer sent me a larger Siemens for the Start/#1 position very shortly.

Original:  Siemens V23073, the 4-pin are -B1005-A302 and the 5-pin is  -B1008-A303

I see no markings indicating surge protection (either "R" or "D"). Certainly the #5 position relay was  not different from the other SPST positions. Only the SPDT in position #1 differs.

I cannot find data sheets for these old relays.

[Speedfrog]

2 hours ago, docc said:

I'm not sure I understand the value of investigating the original relay as it was proved failure prone from mid-2000 year.

Not so much for the value of the original relays but only to get some insight into the original design of the electrical system.

We all know by now the electrical engineeers in Mandello Del Lario had their share of miscalculations, but I'm hopeful they would have made sure sensitive equipment like the ECU was protected from potentially destructive voltage surges coming from said relays.