How to wire AVCS actuator?

[TimH]

Pat,

No, I know they're there for flyback/snubbing, so they won't be dissipating anything of note while the PWM output is being driven - as you know it is when the PWM output turns off that we need to clamp the back EMF: which is more complex to calculate than steady state dissipation.

Cheers,

Tim.

[pat]

Tim,

Just try to think of the whole thing as a buck converter as described here with the notable difference that we are "regulating" (VBAT-Vout) not (Vout-Vground) [actually we want to be and are regulating Icoil]. The diodes will either be reverse biased and therefore dissipating an insignificant amount of power, or forward biased and essentially conducting DC (dI/dt should be small compared to the PWM period ergo current ripple should be low per period). I agree that there will be additional losses during switching transients, BUT they are insignificant. We are not driving the coils anywhere near normal SMPS frequencies, 300 spikes per second is not going to generate remotely close to a 30kHz switching frequency. The vast majority of the power dissipation will be due to forward conduction, which can be measured fairly easily. Nothing complex going on here

Cheers,

Pat.

[TimH]

Thanks Pat - I sometimes have a tendency to over-analyse and over-complicate things in my mind

[TimH]

Related to this...the 2-pin AVCS cam sensors are VR types I believe? Does it matter which pin gets grounded back at the ecu? I'm assuming not since it is simply a pickup coil output?

And if it does matter (why?) which pin is the signal?

[Benkku]

It depends, in terms of the signal polarity. You can test the polarity, with steel bit which is ferromagnetic. Moving bit towards the sensor head area, if voltage goes to positive side and then negative side when moving steel away from sensor head area = correct polarity. Polarity is switched, if voltage goes negative when steel bit is approaching head area and when leaving voltage goes positive.

As what comes to application (AVCS ECU etc.) operation. It might work, with either polarity or not.

[TimH]

But you can choose which edge the Syvecs triggers on so shouldn't matter really?

[pavlo]

Tim,

It does matter in as much as if you want a minimal fuss setup, you will not only have to change the trigger edge but also the angle offsets and the threshold voltages. If you can do it all yourself from scratch this is absolutely not a problem, as soon as you have to work with someone over the phone or something (as I had to recently) it becomes a problem.

So I would advise sticking to convention where you can to avoid unnecessary extra work for you or your customers.

[TimH]

Thanks Paul - all makes sense, and will make sure I do things the "usual" way from now

[Gizmo]

Pat, would you still advocate the use of Schottky diodes on the the S8 or not necessary ?


thanks ---john---

[pavlo]

Yes it's necessary.