MicroSquirt.com read only forum (Jan.15,2018)

Hi All,

I have a problem with RPM dropout at 7500 rpm, engine reline is 11,000rpm

On the datalog the tach count resets and it is like the engine has hit a rev limiter.

What i have tried after reading the forum and receiving some assistance from others.

1) 10k+ Inline resistor gets the motor above a lower 2500 rpm barrier, but there is still the barrier at 7500rpm
2) Seperated VR wires out of the loom and away from the ignition wires and injector wires and away from the Cam trigger wires.
( motor running 36-2 wheel with VR and 1 tooth hall effect cam sensor
3) Now have seperate ground for Cam sensor and VR- to the shield wire and pin 33, but still have the problem
4) I have move the VR sensor away from the trigger wheel by about 1 mm and the rpm drop occured at 4500 rpm , it got worse!
5) VR sensor gap was about 0.9mm , so i closed the gap to about 0.4-0.5mm but this had no effect on the 7500 dropout , but now the required inline resistance to get over the 2500 barrier is only 5k Ohms not the 12-15k ohms needed before.


I have 7 days in which to find a solution before the bike goes to the dyno

Any idea welcome.

Norzilla

Different problem. Norzilla's bike only uses 1 vr. Cam sensor is hall effect. (12v)

I have a very close setup as his. (same oem, diff engine)

I have also seen this just recently with mine. But mine is just above my usual rev limit, and is why i only recently seen it. Any other suggestions are welcome. I will try and pin point it this weekend.

Agreed 800vtwin, I was just going to edit my post & seen your reply.

I should point out that I am using my cased MicroSquirt like a plug n play module, in that I am using female adaptor to the Ampseal plug, which is connected to the standard OEM wiring loom connector, and it has two VR grounds.

No matter what combination in polarity, as soon as I connect the two OEM sensor grounds to the shared ground that the MicroSquirt requires, the engine dies...the diodes stop this happening, and they also allow me to keep the polarity that looks correct in TachRef. Another point is my bike can run without the cam sensor being connected, just needs it for starting, and all my problems have been cam sensor related.

PS FWIW, I looked at Norzillas datalog, and he has TachCount and Engine issues at the affected rpm, so it looks like he has noise or an ignition settings issue at speed. Incidentally, I also notice his vBatt seems high at 18.4V peak, so I'd be tempted to do a datalog with some electrical loading, ie with the lights on etc to lower this, to see if this affects anything, but I suspect it won't.

Hi, re electrical load, the lights were on in this test , not sure if the cooling fans were also on but i can do a datalog today where both are on.

Norzilla wrote:Hi, re electrical load, the lights were on in this test ...

The electrical load isn't the cause of your resets, it'll be noise or a voltage spike at that rpm.

800vtwin wrote: Any other suggestions are welcome.

I would isolate the crank or cam sensor grounds some more, which is essentially what I did, because at the time I didn't realise you couldn't share a common ground with two VR sensors, which is what the MicroSquirt loom is pushing you to do...which is why the polarity issue was so important apparently. If I'd known the cased MicroSquirt wasn't designed to take two VR sensors on VR1 & VR2, I would have used VR conditioning boards from the start, especially on my cam sensor.

I noticed you had 2 vr's and why i posted different problem in my first post. As we have 1 vr and 1 hall effect. The hall effect is grounded to sensor ground, and is seperate from the vr.

That being said. Norzilla just took your advice, and installed a diode in series on his cam sensor ground. (power supply to sensor). His problem was cured. (i'm not sure why yet?)

The high battery voltage is from a defective voltage regulator, which he realized after. At first i thought the diode was just dropping the voltage to the cam sensor, therefore fixing the over voltage supply to it. But he unplugged the reg and retested for me, and without the diode and system voltage back to norm, it still dropped tach count at 7500. The diode again fixed this even with normal system voltage. So it's not the forward voltage drop.

So my theory is the diode is somehow filtering noise. From what i know about diodes, this doesn't make sense. can a diode filter noise? A capacitor would be a better option if its noise here. The only thing i can think of is maybe there is ac noise in the system ground, and being stopped on return from the cam sensor. I do believe hall effects will produce some internal noise themselves, and this may play a part? And ac noise isn't stopped on the way to the sensor, its stopped only from returning back to the ground.

I'm trying to grasp whats happening here? Any suggestions?

Anybody an electonics guru?

A VR puts out an ac voltage.(sine wave) Running them together allows the voltage pulses on both lines. It makes perfect sense for you to add a diode to stop the ac return. Its perfectly fine to use 2 vr's. But you must seperate them, as you did. Both vr wires carry the signal. You can't hook 2 signals together.

A hall effect is a dc signal. 2 wires (pos/ground) power the sensor - 1 wire outputs a square wave signal. The power supply has nothing to do with the signal. So you can have the same power supply on all sensors. (run 5 sensors hooked together) and the signal outputs are still separated.

You are fine with 2 vr's and diodes. But its not a noise issue, but crossed signals.

I believe a vr is more accurate and faster than a hall effect. But the vr is more suseptable to noise. And this why the use of the 2 different sensors and where they are used. The cleaner hall effect is used once per rev but the vr is used 72 times per rev. (on a 36 wheel eg) Someone correct this if i'm wrong.



Anybody know whats happening with the diode in hall effect power supply??????

Hi

I want to know of you solve your rpm droout problem, I install a microsquirt v3 on an Alfa 146 with 60-2 configuration , but the engine don't rise over 4000 rpm, then I install a 10k resistor then the engine rise to 5000 rpms. I install a 5k resistor beteween + - vr terminals and the engine run 6000 rpm but the engine don't want to go over more.

please tell me if you have a solution

JC

Hi
OK, which version of the MicroSquirt are you using?
This is a very old post, but when you say

and the engine run 6000 rpm but the engine don't want to go over more.

Is this it won't start? If you load up the VR signal too much, there is not enough voltage to drive the circuits. With the older v1 & v2 cased MicroSquirts the cam sensor circuit (VR2 IN) needed 1.8V peak to peak AC to register.

I would strongly recommend conditioning the VR circuits with jbperf.com small board. It uses the same chip as the MicroSquirt v3, and works very well.

If you have access to an oscilloscope, you can see the results of your resistors. Maybe you can heck the peak to peak voltage with the plugs out on cranking with the starter.
You can also change the distance, increasing the distance reduces the voltage too, but how high do you want to rev?

I would suggest that you just need to tweak the resistance some more, nearly all my problems were on the VR2 circuit and the common ground which is shared by VR1 circuit, and as the VR1 signal was stronger it would interfere with the VR1 signal and confuse the MicroSquirt...resulting in a misfire/rev limit situation.

Good luck
Mike