Current stops first?

picowit · Post by **picowit** » Sun Oct 11, 2009 2:06 pm

I am at present investigating the opening time of automotive fuel injectors. The attached pic shows 3 traces:

Blue: the control signal (Gate) of the transistor that control the injector
Red: the injector current through a 0.05Ohm resistor connected to ground
Brown: the voltage over the transistor that controls the injector (not relevant to this question)

1) How is it that the current stops flowing (red line starts dropping) BEFORE the transistor switches off (blue line drops to 0V)? (It also looks like current starts flowing BEFORE the tranny switches on...)

2) Why is the rising edge of the blue pulse delayed by a tiny amount with respect to the yellow triangle?

Erik · Post by **Erik** » Mon Oct 12, 2009 8:24 am

If you are sampling at the maximum sampling rate of some of our devices the samples for the different channels are not taken at the same time, they are taken sequentially.

This could give the impression that the signals are not aligned.
You could try swapping the channels around to prove that this is the case.

picowit · Post by **picowit** » Mon Oct 12, 2009 8:50 am

Erik wrote:If you are sampling at the maximum sampling rate of some of our devices the samples for the different channels are not taken at the same time, they are taken sequentially.

This could give the impression that the signals are not aligned.
You could try swapping the channels around to prove that this is the case.

Thanks for the reply!

I thought of that and did it and got the same result. I still remember it took me a while to get the scope to trigger on the B channel. But I'll do it again - just to be sure and will take screen shots and post.

Will also reduce sampling rate to see if I can find the point at which things make sense.

picowit · Post by **picowit** » Mon Oct 12, 2009 9:16 am

OK, it seems like this situation actually has more to do with the amount of low pass filtering (to get a nice clean signal) that is active, than the sampling rate.

Only when I reduced the sampling rate to 2KS did the signals match.

I then increased the sampling rate again and correspondingly increased the filtering and was able to get matching signals up to 2MS.

Increasing the filtering seems to INCREASE the amount by which the current signal LEADS the control signal.

Why the low pass filtering should make the current signal appear to occur BEFORE the control signal is still a mystery. It should be lagging should it not? And the lag should INcrease as the filter frequency is reduced surely?

... and then Picoscope 6.2.14 hung again (with "Capturing" appearing bottom right with a moving blue bar )