Let's take a look at a fuel injector tester and indeed, a fuel injector. This is a device that's used in vehicle engines and you've basically got a high pressure fuel inlet at this end and a small filter and then you get a very fine nozzle at this end that puts out a speed like an aerosol and the inside is a solenoid valve. and if I hook this up to a power supply and give it a jab, you'll hear it clicking. I'm not going to hold it up to the microphone again like I did in the previous video when it really made distorted pops, but anyway, it's going to click.

Hopefully you can hear it clicking and in operation it just uh basically speaking when it needs fuel in that cylinder, it just gives it a short pulse and it sprays a a controlled portion of fuel in this little Teamster is designed for testing them comes as instructions in glorious Bill page Chinese and also in English Makes me think this is probably used a lot in Chinese garages. I Thought this was going to be super simple. it should be super simple, but something weird happened I shall connect this up and initially when I got this, I connected to a 12 volt Supply and it didn't trigger the uh, the fuel injector. Oh, Incidentally, for technical reference, the fuel injector coil is about 50 ohms impedance or resistance and it draws about 800 milliamps at 12 volts.

Let me bring in the power supply here set to just above 12 volts because it doesn't work if you put it to 12 volts. That caught me out. it just didn't work at all. It lit the little LED and looked as though it was controlling the unit, but the unit wasn't triggering.

But now I've done that. That's uh, that sleeve has fallen off there again. Now I've hooked that up. You choose the mode you want.

The first mode: I'll Zoom down this just a little bit. The first mode fires a single pulse. the LED lights. You can hear that click.

The second mode you'd think you could just work through the modes but know you have to press reset every time you start to scratch. The software is extremely primitive. The next does a sequence of pulses, then the next mode one two three does a longer sequence of shorter poses, and then the next mode pulses it continually at a low speed. Okay, so now we've seen it working.

Well, we've not seen it working I've not shown you it's spraying liquid out. Um, that would involve an adapter and a high pressure sort of liquid roll here. a solidwork. You could make an aerosol automated air freshener.

this if you could, get an adapter. Unfortunately, it's not so easier to adapt that to the inside core. It's got the manifold that pushes into the high pressure fuel manifold, has a an outer o-ring here for fitting on. Okay, let's pop it open.

Four screws in the back and what will it reveal? So there's the incoming Supply So there's a what looks like a Polarity Protection Diode a 7 8 L05, which is the 5 volt. Supply Probably for the microcontroller, there's a display, possibly just a single resistor for the whole display, but it would have been nice to have one resistor pair LED There may actually be a resistor pair LED That's good, but very simple to drive that single digit display. Uh, that's probably the mosfet for switching the output here. And then there's a 7A Mo file which is a 5 volt regulator which is odd, right? tell you what, and that's probably prior to the back.

EMF Protection node I'm just going to reverse engineer this one moment, please. Okay, let's explore. This thing was spill of so many surprises that I had to crack out an oscilloscope. Just a simple little single Channel oscilloscope.

Just a moment to the output waveform to answer some questions. That oscilloscope bit is at the end of the video and uh, it did answer the questions. Let's Zoom down on this image a little bit, not too much and explore it. So what do we have here? We've got the supply communion and the positive goes via the Priority Protection Diode.

It goes to this small 78l05 regulator that powers the microcontroller, but there's some odd circuitry. There's a transistor here that can turn that microcontroller on and off, then the zip file that incoming Supply goes over to the next five volt regulator 70 m05. but it's not always being used to regulate at 5 volts, it's also majorly current limiting. The look of it, it's certainly causing a significant voltage drop.

It's making me wonder if that was a great choice for this. but it is what they used and that has its negative connection. The Ground Zero Volt Whatever you want to call it connection that would normally be just tied to the zero volt reel. it has it sitting on top of a Zener diode.

It says 6.8 volt here. If that's the case, it would be I'm wonder if that is 6.8 volt. Wonder if that's why I was capping the voltage? That'd be interesting to see if it's just been dropped across that, but um, the regulator can either sit on top of that Zener diode which raises the 5 volts above that level, giving theoretically about 12 volts. But when this transistor turns on and it shunts that out, it then ties the zero volt of the Reel of the regulator down to actual zero volts and then it just puts out 5 volts and it looks by the oscilloscope traces.

if that's to cut the power dissipation from the coil by initially stars at higher voltage and then dropping the voltage done which really needed, but they did it that way. The variable voltage is fed down to one end of the injector coil and the other end is Switched via this mosfet. Now it does say Mjd127g which I believe maybe a PNP Darlington which would be a completely unsuitable transistor for this location, but they've used a mosfet which is a much more suitable one. and uh, there is that clamp diode across the coil for shunting the back EMF specs of three wheelish type diode.

The display: There's some same display just uses seven segments. It's got a common negative and then it's got seven resistors. One big one over here just to jump track. So though to be honest, they didn't really need to do that, they could have rooted the tracks differently for this.

I'm not sure why they did it that way, because uh, there's plenty of pins to make a controller and uh, it's just unusual they did it that way. They didn't need that, but but they did. There's lots of things about this that make me think it's got stolen software in it that they've just repurposed. Particularly, it seems quite old-fashioned the way it does things okay.

bring in the schematic for exploration. Here is the schematic and we'll zoom in a little bit more and see if I can do it in a controlled manner. This time we have the incoming Supply going through the diode to create a roughly 12-ish volt rail that goes to the five volt regulator which Powers the microcontroller via this transistor. the transistor is a PNP transistor so it's switching to the positive reel and to turn it on, you pull it down to the zero Volt Rail and in this case they've used Uh one key resistor pull that down to keep it turned on.

But when you press the reset button to get out of the margarine, which is the software is just bizarrely primitive. When you press that, it puts 100 ohm resistor from there up to the Uh Five Volt Rail and that effectively turns that transistor off and that turns the whole processor off. When the processor is turned on though, initially this capacitor will charge up via this resistor. Meaning, it starts off around about zero volts and it charges up to five.

That's the reset input and that just basically provides a clean reset. There's an oscillator pin being used with an external resistor. The data sheet for this chip doesn't show this capacitor. they've added that for some reason, but it just shows a resistor to the zero volt rail for fixing a speed and that's a 75 key resistor.

They've just chosen it for the timing. It makes me think this could have used a Pic maker controller type thing with built-in oscillator, but they didn't do that. It could have also used one with the built-in reset circuitry and it does have a watchdog. Um, that's one thing in its favor.

A watchdog is a counter and timer inside the microcontroller that at the beginning of each Loop of your software, you just reset it. And what happens is that if the software is running normally, it just keeps resetting it back to zero. But if the software locks up in any way and doesn't get back to the start of the Rope, it means that time will time out because it's not been reset and it will reset the processor and reboot it. There is the seven Cent display connected to the zero Volt.

Reel With the seven resistors I've just drawn one. for clarity, two buttons to the zero volt reel for the mode selection, mode set, and pulse. So strange, they shouldn't even need the reset. They should just need mode and you should be able to just switch through mode at any time.

It would not have been hard to implement. Uh. Moving on to the weird secretary for driving the injector. The injector colored orange here.

It's a bit of a cramped drawing with lots of jumping about here just because it is a complicated circuit. The supply comes to. the 5 volt regulator has a decoupling capacitor there and one in the output to the its zero volt reference, but that normally sits on this Zener diode and that lifts that up. So the output of that five volt regulator is 5 volts plus the voltage across that Zener diode that supplies power to one end of the injector.

The injector has the diode across it for the shunting the back you may have Spike and there is the mosfet switched by that 2K resistor to the zero volt rail. There is the LED and a 510 Ohm resistor that's a little green LED under there. and uh, it is powered from street from the 12 volt rail just to zero volts via the mosfet. So whenever that turns on, regardless of the voltage there, it's just going to let that Led at the same intensity that is more or less it.

So what's actually happening this situation is that it turns out that this mosfet is being turned on while this transistor is turned off and that gives that extra voltage boost. So the coilers initially brought in with that sort of Spike of higher voltage and then just after that very short time after that, this transistor turns on and it shunts this mosfet and that then takes the five volt regulator down to the zero volt reel and that then just puts out the 5 volts and that will reduce the dissipation across that. But the best way to show you this is to show you it with the oscilloscope one moment please. The experiment is set up and we do have our answer.

Instead of using the injector, I've used a more convenient load a 10 Ohm resistor with the clips on it and even running in Mode 4. it's uh, which is continuous. it's not getting too hot for I think this is a one watt carbon film resistor and if I dim the lights and take the exposure off so we can actually see the display here and I'll Zoom down that that would be a good idea too. So here is the display.

We're seeing that it actually goes up to eight volts. It doesn't go up to the field 12 volts. That voltage it goes to depends on the load, but even open circuit it only went to 10 volts. There's a significant voltage drop across the circuitry in this unit over here, but it goes in the case of having a decent load connected.

It goes up to 8 volts initially to in the case of the injector to give the coil a good boost to start the magnetic attraction of the core and then just after that it activates the auxiliary transistor that bypasses the Zener diode, under shunts the ground connection of the 78 M05 regulator to the zero volt reel, and that drops it down to 5 volts. So that is what it's doing. So watch your eyes, the light is about to come back. The light is back.

That answers the question. What a strange thing. Do actual car systems at fuel control systems do that as well? Do they switch between the two voltages I would have thought it would just providing it very distinct length pulses. Did they feel that they had to do that for lowering dissipation? Maybe some do that, maybe some don't.

I'm not really sure the true Aficionado Mechanics: The experts may actually know more about this than me. Eric All comes to mind with those oscilloscopes. That would have been interesting. Um, to see what he thought of that.

But there we go. That's an interesting unit. The circuitry is much more complex, but also weirder than I Was expecting. particularly that reset mode.

um, that basically just shuts the processor off and starts it again. But there we have it. The more complicated than expected and strangely hobbled as well. Um, cheap.

eBay Fuel injector tester.