A unique starter relay problem and solution

Now that I have something potentially useful to share, lets start with a quick intro...

I am new to the forum and Aprilias for that matter, picked a 2000 RSV mille a couple of weeks ago. I cannot get enough of it, these things are so much fun!


Almost quit riding all together, it was just not as fun as it used to be...
Although now I can say for certain, suzuki is to thank for that and what they call the sv1000s. Now that was a miserable motorcycle...
It was not fun, it was not reliable, caused all sorts of cramps and aches... (goes on and on)
Well enough of that.


The previous owner warned me about a starter relay problem, sometimes it would not turn on the starter motor. He suggested I give it a good smack to get it going again. Ran without a problem for a good little while, although the bike would simply refuse to start in gear (regardless of the position of the clutch lever).

The other day, I experienced what he was talking about... But something did not seem right. With the ignition key turned to on, all the lights would go on and the fuel pump would prime etc.
But push the starter button and nothing but the fuel pump whine.
Wiggle things around a little bit and it would start, odd...

After doing some research I discovered that the starter relay is not a stock one, but it appears to be a Porsche one. Also ran some tests on it, and it performed beautifully. So that was not the cause.

Further investigation revealed that the starter relay was only getting 7.5V across its terminals when the starter button was pressed. Which meant that there was a 5.3V drop somewhere in the starting circuit!

Performed a quick current test, and a problem suddenly became clear...
The bloody thing was drawing a massive 1.9A!!!
There is a reason automotive relays should not be used.
Did a quick calculation to figure out what the resistance was in the starting circuit, and it was about 2.8 Ohms (which I believe to be within reason for a circuit passing through half a dozen connectors and small switches).

Although cleaning and sealing the connectors would not hurt, that is just far too much current to be passing through that circuit.
The higher the current the greater the voltage drop...
So the 2.8 Ohms leads to a voltage drop high enough for the start signal to fall below the starter relays threshold.

The simple fix would be to replace the Porsche relay with the Yamaha unit, but due to a lack of funds... I had to get a little creative.

What I decided to do is to eliminate the starter relay from the circuit, and in its place put a driving relay. Which would then drive the starter relay from a fused circuit attached directly to the battery.



This way I do not spend an excessive amount of money, and have a happy starting circuit. At first I decided upon a 12V/40A automotive relay, which has a coil current of 0.133A (much better than 1.9A!!).
Which would equate to a voltage drop of a mere 0.4V!

While at the store I decided to take a risk, and try a non-automotive relay the reason being there is no need for a 40A relay to drive the starter relay as it "only" uses about 2A. So I went with a standard relay rated for 5A, I just hope it will withstand the vibration (its rated for 10-50Hz with a 2.5mm double amplitude).

So I put everything together... and everything works!
Not to mention now I can start the bike in gear.






Not the most elegant layout, but at least all the connections are crimped, soldered, and shrink wrapped. Hopefully it all soon be replaced.

Conclusion:
Do NOT use a automotive relay unless you want to make a driver for it!
Also just doing some power calculations...
Before the starting circuit had to dissipate 10W
Now the starting circuit has to only dissipate 10mW

That is THREE orders of magnitude less!

That should eliminate a great amount of stress from the connectors, buttons, and the diode block... So all those components are far more likely to last much longer.

But this too is a temporary solution, once I get some money it would be nice to replace the starter relay with a solid state one... which would mean that it draws even less current from the starting circuit and will NEVER fail.

Well hope this is helpful to some...

Thank you for the warm welcome!

I will make sure to post updates...
This morning everything fired right up, and it would have been a good test too.
The weather has been quite odd lately, low 10s at night and high 30s in the day...
a good test for condensation
(this may also have been the reason the problem reared its ugly head in the first place)

On another note I have been looking at Solid State Relays (SSR), and have come up with some potential candidates...


For an all-in-one solution we have the following:

Teledyne - S75DC150
From a functional perspective this one is the BEST, unfortunately it is hard to find and is rather expensive (~$280). **Apparently its a new unit, and the cost should come down!**



Crydom - D06D100 or D06D80
The 100A and 80A models, these should do the trick... But I would be concerned about using the 80A model as the parts book calls for a 100A unit. These are $140 and $120 respectively, and seem to be far more common. With an input current under 28mA, and output resistance of 5mOhm and 8mOhm. Since their duty cycle would be under 5% (that is a "worst" case scenario where the starter needs to be run for 30s), I don't foresee problems with overheating. Leak current, which is a problem with SSRs, is between 0.1~0.2mA which is insignificant.



Another option would be to replace the driving relay with an SSR, the following would work and are far more cost effective:

Crydom - DC60S5
This is a big sucker, same form factor as the all-in-one unit above. Fortunately its cheap ($21), and meets are the requirements rather well.



Crydom - CMX60D5
This one is tiny, say about the size of a large postage stamp. There are two downsides to this one, it is about $24 and will require an input voltage limiter. The reason for this is because its input range is 3-10V, but a simple shunt should do it or even a resistor.



So we shall see what will happen... As for the next project, I need bigger and brighter turn signals. I am thinking of picking up some original turn signals and converting them to LED. The trick with this one will be to get rid of the thermal blinker relay, in order for it to work correctly I would need to add resistors which would do nothing in terms of power conservation. So far I am thinking 555 timer with either MOSFET or SSR drive.

I am very curious how the SSR works out...
Also I was thinking about where to mount the suppression diode, motor side would be most logical (perhaps in the form of a "wire" going from the motor to the engine for ground?).

Another update, well there have been no problems what so ever with the mini relay. Nice positive starts every time.

On another note, I figured out how to convince the flasher relay to accept LEDs without external resistor blocks (which negate part of the point of LEDs, which is to conserve power)... So if anyone is interested I can do a quick write up.
Although I am not sure if Aprilias use the same relay as Ducatis. If its fully mechanical... well I am working on that one too... Should be coming to life soon.

300miles later, and two days of rather heavy rain... The Aprilia has been performing admirably, with the only glitch being a sticky front brake light switch.
So it would be safe to say that those cheap little relays aren't too afraid of water.

Well I spoke too soon...
Third day of rain did it in, oh well.
Picked up a water and vibration "proof" automotive relay.

Did a quick swap to a 30A automotive relay this morning, this was the lowest rating that they carried (definitely overkill)...


No where near as "good" as the little one with 2.5W power consumption vs 10mW
but there is no reason for it go nuts like the other one and stick.
*The draw is low enough where I can start the bike in gear with the clutch pulled in*

I also tore apart the little one to see what could have possibly gone wrong...


There is less than 0.5mm clearance between the contacts, and there is evidence of sticking at the contact points...