The problem with holding any injector open continually is HEAT. The power needed to quickly open an injector for a very short portion of the cycle time is far more than it can handle in a steady state mode.

Yes that is 100% correct, but I think Alec was asking why we couldn't just pulse the injector open, and then let it close on it's own (if it can) like a typical fuel injector. The injector would still be pulsed (not held wide open)

 

I clearly don't understand the operating principle of your cleaning machine. What is the problem with the injector staying open?
As for the scope shot, and in view of uncertainty as to what it is showing, if the wider pulse is the one for opening the injector, could the narrower one be merely a back-emf pulse generated inductively when the other pulse terminates and the actuator coil is de-energised?

I think you are referring to something like this. The picture below is a typical (non-eTec) fuel injector. You can see the pulse is applied, and then you get the reverse pulse (I know I'm saying that wrong) when the signal is terminated. I think this is the little spike that you see on the eTec injector signal that I scoped.

 

My reason for wanting to make this "new" box is because with my current equipment I can only run 1 injector at a time.

So where are we? If you start from scratch, you need components to generate the signal. Not a big deal, but it takes time to design and build even a simple circuit. Using the existing signal and just amplifying it to drive multiple injectors is another option. Now that I say it out loud though, pulsing them all at the same instant would require a LOT of current. It might make sense to pulse all the injectors sequentially. Same average current but much lower instantaneous peak current.

 

So where are we? If you start from scratch, you need components to generate the signal. Not a big deal, but it takes time to design and build even a simple circuit. Using the existing signal and just amplifying it to drive multiple injectors is another option. Now that I say it out loud though, pulsing them all at the same instant would require a LOT of current. It might make sense to pulse all the injectors sequentially. Same average current but much lower instantaneous peak current.

My preference would be to have a stand-alone box for this task, so I can keep using my current equipment for other tasks while these clean. My equipment is relatively expensive (~$5000 for the "eTec" box, plus ~$15,000 for my main box which powers the eTec box) and things are expanding here at a rapid rate. If I use the existing equipment to power my new 6-injector box, chances are good that I will need to buy an additional machine in the near future. What do you think that would take to achieve?

As far as firing them sequentially I completely agree. I wouldn't expect to or require to fire them all at once in any scenario.

 

I think part of the confusion is the screen on your scope is wrapping. It's triggering near the center, then the negative pulse on the left of the screen is actually after the screen has wrapped around.

Try this to get a clearer mental picture for what's happening. Keep your trigger on rising edge, but move the trigger point almost all the way left on the screen. Lengthen the time base so that you can fit multiple complete cycles on the screen. Now shorten the time base until you have one complete cycle on the screen. This will give you a clearer picture of what's happening.

Also if you have 2 probes, and assuming the signal to the injector could actually be inverting, and assuming the injector has 2 wires going into it, you may get a better result by attaching one probe to each of the 2 wires going into the injector, and connect the ground clips to the motor chassis. Now set your scope to show the difference between the two probes, sometimes shown as A-B in the math section of the scope settings. This gives the same effect as a differential probe and can give you a more accurate reading if neither of the 2 wires going into the injector are common to the motor ground.

 

Ok, I will re-try the test again tonight. I have not really used the scope much at all so some things are pretty new to me.

 

OK, so now we know a bit more about the application. So what you need is a pair of 555 timers, or a 556 dual timer, or else a way to set the duty cycle for on and off time. Then, to make it universal, an adjustable DC supply with enough current to handle the biggest injector you anticipate , as well as enough voltage to drive it. Then, since you say sequential driving is OK, a sequencer, possibly like a CD4017 counter, or the similar octal counter. plus and gates to make it simple to distribute the pulses around the ring. Nothing terribly complex except the mechanical arrangement to assure that the whole system is very reliable.

 

OK, so now we know a bit more about the application. So what you need is a pair of 555 timers, or a 556 dual timer, or else a way to set the duty cycle for on and off time. Then, to make it universal, an adjustable DC supply with enough current to handle the biggest injector you anticipate , as well as enough voltage to drive it. Then, since you say sequential driving is OK, a sequencer, possibly like a CD4017 counter, or the similar octal counter. plus and gates to make it simple to distribute the pulses around the ring. Nothing terribly complex except the mechanical arrangement to assure that the whole system is very reliable.

Ok great! I think we're really getting somewhere. As far as the DC power supply, I think I'm going to go with the same supply that the current box is using. If it is good enough to power 1 eTec injector it should be fine for this use (powering 1 at a time). I like the adjustable idea though, that way I could power regular 12v injectors as well if I need to. I guess we would have to turn off the reverse pulse for that as well though.

 

So this would make it 54V,

So where on a outboard motor does the 54V come from?

 

I haven't read the details on these injectors but are they piezo injectors?

The link to them makes it look like it is using some sort of "voice coil" movement. I say that from the moving coil around a magnet.

 

So where on a outboard motor does the 54V come from?

55V is supplied from the stator.

The link to them makes it look like it is using some sort of "voice coil" movement. I say that from the moving coil around a magnet.

Yes, there are a voice coil. I can't seem to find it, but I have seen that in writing somewhere, in fact it might be on my website.
https://injectorservice.com/evinrude-etec-fuel-system

EDIT: The predecessor to the eTec, which is the Ficht, was more like a conventional injector, that used a mechanical spring to do the complete pintle closing process. They didn't work very well and eventually OMC went bankrupt, and BRP bought them out. After the buyout the system was re-designed and that is where we get the eTec from.

 

It makes pretty good sense to minimize the spring if you think about it. You still need it to ensure the power-off position but otherwise you can do more with electronics than with a spring. You waste less electricity by not trying to overcome a strong spring. That means the coil and magnets can be smaller. More energy into pumping the fuel.

 

So where on a outboard motor does the 54V come from?

That big pulse may come from a magneto Outboards often have magnetos to avoid needing battery power for ignition.
And, for the cleaning process, it seems reasonable to forget about that reverse pulse that is used to quickly shut off the fuel flow. Just an on pulse to open it and then let the spring shut it slowly. And there must be some spring to prevent the injector from flowing when the engine is not running. A leaking injector could fill a cylinder with fuel, causing all kinds of nasty grief.

 

I would think that after cleaning there would be some testing, or at least the option to do some testing (maybe in the future if not now?), in which case you would want to replicate the signal from the motor to at least the degree that you could verify behavior. Meaning if it were me making it, I would try to implement both pulses.

 

I would think that after cleaning there would be some testing, or at least the option to do some testing (maybe in the future if not now?), in which case you would want to replicate the signal from the motor to at least the degree that you could verify behavior. Meaning if it were me making it, I would try to implement both pulses.

The TS already owns a tester. We are discussing a separate system to be used when cleaning the injectors.

 

Ok, so I'm starting to put together a parts list. Below is the power supply that they are using in the 55v eTec box. If I want to go adjustable from 55 to 12v, I will need to get an adjustable box, but maybe for now I should work on making a dedicated 55V box.

2 - 556 or 555 timers - how do I go about choosing which one I need there are lots to choose from. - This is the piece that will set the signal duty cycle, correct?

1- CD4017 Counter - I assume this is what is going to send the signal to each injector individually?

https://www.digikey.ca/product-detail/en/cui-inc/PDRA-240-48/102-4965-ND/9555718

 

The TS already owns a tester. We are discussing a separate system to be used when cleaning the injectors.

This is correct. I already have a commercially made tester AND cleaner. All in one. But its limited to doing a single eTec injector at a time. I'm trying to cut out a bunch of time on cleaning by cleaning 6 at a time.

 

Ok, so I'm starting to put together a parts list. Below is the power supply that they are using in the 55v eTec box. If I want to go adjustable from 55 to 12v, I will need to get an adjustable box, but maybe for now I should work on making a dedicated 55V box.

2 - 556 or 555 timers - how do I go about choosing which one I need there are lots to choose from. - This is the piece that will set the signal duty cycle, correct?
Right. It would probably be a 555 timer IC using a circuit that allows adjustment of both the frequency and the duty cycle. For a cleaning operation probably just one frequency and one suitable duty cycle will be enough. Possibly as much as a 2% on time is what you could get away with, but maybe only 1%.
1- CD4017 Counter - I assume this is what is going to send the signal to each injector individually?
The counter will need to have it's output gated with and gates so that the short duty cycle can be applied. Otherwise you would have a ten percent on time for each injector and that is probably way to much on time and heating.
But since these are logic-level devices each output will need a driver circuit as well. I do not recall any comment about the required current but I am guessing thatit is several amps, and so the switching transistor will need to have that capability, and an adequate voltage capability as well. So those may be some $5 transistors. Each output line will also need a spike suppression diode, but they can all share one common current limiting resistor since only one at a time is pulsed.
The big expense will be the heat sinks and the injector connectors, other than that it will not be a terribly expensive addition. And it will need a small 12 volt supply for the logic devices.
https://www.digikey.ca/product-detail/en/cui-inc/PDRA-240-48/102-4965-ND/9555718

 

Can you monitor the injector current during the cleaning process? (You might need to insert a current-sense resistor in series with the injector to do that).
Since you can monitor the existing cleaning pulse voltage, why not use that voltage, instead of separate timers, to trigger/control the sequential pulse timing for multiple injectors?

 

Yes, there are a voice coil. I can't seem to find it, but I have seen that in writing somewhere

I saw it in the link Alec gave in post #3.

And, for the cleaning process, it seems reasonable to forget about that reverse pulse that is used to quickly shut off the fuel flow

If you look at how they are made and work(in the illustration above) both pulses are needed to clean the whole injector, it works as a pump too.