Need some help from someone smarter than I am....Solid state relay wiring help

Thread Starter

wackerjr

Joined Feb 10, 2010
3
Hello new to the forum here but I think I am in the right place due to my searches :)

I need some help, I am currently trying to figure out how I can use a solid state relay in an automotive aplication to switch polarity...

I have a data logger hooked up to my vehicle and am trying to log the injecter pulsewidth. However my injectors are fired by ground... as in the injector is fed a 120v constant, and the pcm tells which injector to be fired by the proper ground being activated low....

my problem is that My logger only has an input to read High side inputs..... I can use a Electromagnetic relay but I don't think they will respond quick enough for accurate measurement... in come solid state.. I am trying to record down to 1.5ms aprox, so I need something fast to be up to the task....

But here is my question is this possible? If so how would you guys go about wiring said SSR to accomplish the task of showing a 12v reference being activated by the low side of the injector?


any help would be appreciated.... Thanks
 

Thread Starter

wackerjr

Joined Feb 10, 2010
3
I think I need to word it differently....

I have an injector with the 120vdc side constant to a large electromagnet, the injector is energized by the pcm sending a ground to the injector to energize the magnet.....as long as the pcm is sending ground the inj is energized.

my logger has a 12v input....

I need to monitor the ground side of the injector (I actually am watching how long this ground is "on" per se', this gives me the pulsewidth of the injector)

so in essence I need to monitor a ground reference, with a voltage input on the logger (switching polarity from - to + 12vdc) I usually do this with a standard bosch type relay, however I think a elctromagnetic relay is going to be to slow to perform the task.....

Your thoughts... I thought just a simple SSR wired to the 140v injector for the input, and on the output side sending it to the logger reference input.....

Discuss :)
 

beenthere

Joined Apr 20, 2004
15,819
A SSR tends to latch on with DC applied.

Why do you need to switch a polarity? What does it take for your datalogger to sense an event?
 

MikeML

Joined Oct 2, 2009
5,444
Suppose you want to measure this with an oscilloscope. How would you do it?
Chances are, you would use a 10X probe, which will constain the 120V signal to only 12V. If the signal swings both positive and negative, then you would use the "level shift" knob on the scope to move the trace to the center of the screen so that you can observe both positive and negative voltages.

In the context of the scope, what can your logger not do? Can it log a voltage from -12V to +12V, or is it constrained just to 0 to +12V?

What is the input impedance of your logger? That will determine what value of resistor will be reguired to make a 10X scope probe equivalent for your logger.
 

Thread Starter

wackerjr

Joined Feb 10, 2010
3
the logger can only do 0-12v..... I wish it did -12-12v then I wouldnt have this problem :) signal is only negative/ground in this case

the injector has a constant 120v to it, it only fires when the ground is sent through the pcm.....

Basically I can't monitor ground, only voltage on the logger...0-12v

I have in the past just used a relay to switch polarity, but in this case we are talking about 1-4ms of time here, and I think a standard relay won't work....

I am open for other ideas, the relay is just what I know :)

on a side note the signal from the injector on the ground side is very dirty, my thoughts with the relay were to also clean up the signal a bit for the logger as well.....
 

MikeML

Joined Oct 2, 2009
5,444
Ok, Here is what I suggest. I have a USB plug-in data logger for my laptop computer, and although it accepts +- voltages, the maximum voltage is limited to +-20V. I have successfully used it to record the voltage across the points of aircraft magnetos, which have peak voltages of several hundred volts. I did something similar to what follows:

I cobbled up a simulation of what I think your injector circuit is like. It starts with a 120V power supply, has an inductive coil, with some sort of snubber/capacitance. It is switched to ground by a control signal from the PCM. I'm guessing that the switch is an NFET, but it really doesn't matter for the purpose of this discussion.

Look at the Blue trace in the simulation. It is the voltage (that you want to record) at the bottom end of the injector coil V(Injector). Note that it swings from 120V to near ground, and then it oscillates wildly up to several hundred volts as the switch turns off. Obviously, you do not want to connect this to your data logger directly.

I created an Attenuator and Offset circuit to put between the injector voltage and the input of your logger. It does three things: 1. it uses a resistive voltage divider to attenuate the signal so that it will fit within the allowed range for the logger. 2. It uses a common 9V Alkaline battery to offset the attenuated signal by +2V, allowing you to record voltages that swing below ground. 3. It uses the two diodes to clamp the signal if it tries to go below the logger's ground, or above the 9V battery voltage, preventing destruction of the logger if some huge spike comes in the input.

The red trace shows what the Logger would see. Note that to interpret the signal, you first subtract 2V from every Logged value, and then multiply by 20. For example, before the NFET turns on, the Blue line is at 120V. The Logger sees 8V. 8-2 = 6, 6*20 = 120. After the Fet turns on, the Logger voltage is 2V, so 2-2 = 0, which is where the Blue line is...

Note that the big positive peaks got removed from the Logger signal by the clamping action of the upper diode. You will need a fresh 9V battery, some battery clips, three resistors, and two diodes to construct the attenuator/offset/clamp.
 

Attachments

Top