Home Made Fuel Injector Cleaning Machine

[robertisaar]

They are essentially connected on the outside, but that doesn't necessarily mean there isn't 2 drivers inside the ECM. What is the other pin connected to on the board? It has to connected to something...

i'm saying i think they are bridged INSIDE the ECM as well... it's been a while since i cracked one open though... the ludis recreation docs would probably show it though.

AFAIK, the 1227749 is one of the few MPFI ECMs that uses two drivers.

EDIT: here we go:

http://www.exatorq.com/ludis_obd1/p4xref.html#1227749

"1227730 w/ one missing quad driver, w/ additional injector driver FET"

so either the 7749 has 3 injector drivers and the 7730 has 2, or the 7749 has 2 and the 7730 has 1.

[bszopi]

They are essentially connected on the outside, but that doesn't necessarily mean there isn't 2 drivers inside the ECM. What is the other pin connected to on the board? It has to connected to something...

[robertisaar]

In a typical GM ECM, you either have 2 (batch fire) or 6 (sequential fire) injector drivers. So, it would only make sense in trying to create a controller that you would need multiple drivers to run all the injectors.

IIRC, the 7727 and 7730 type ECMs have 1 driver... the second pin isn't connected to a second driver, just bridged from the first.

[bszopi]

In a typical GM ECM, you either have 2 (batch fire) or 6 (sequential fire) injector drivers. So, it would only make sense in trying to create a controller that you would need multiple drivers to run all the injectors.

[Tbay99Venture]

Well after providing Robert I with some useless advice back on the ALDL thread, I was debating whether or not to post this. But here's a link of a commercial chip meant for operating injectors. They give a sample diagram with transistor specs and so forth, that's not the problem.

http://www.national.com/ds/LM/LM1949.pdf

What this shows is you end up getting into a bit of electronics fabrication to accomplish what everyone hoped would be just a simple fan-out from one injector to 6. You have to find a supplier, build up a circuit board and so on -- not a show stopper but it's cutting down on the potential number of folks willing to do so.

Now I say that the controller approach is needed since you're trying to operate the injectors as they would go in a real engine. A current pulse much higher than the device can sustain is required to get them opening quickly, and then you drop to a steady hold-open current. If you string all of the devices in parallel the feedback to the controller isn't going to be right even if you have a transistor (or bank of them) able to drive them. So it looks like there isn't a short cut to parallel 6 of them. You might be able to get clever and run two or 3 per driver chip and cut down on the parts count though.

Food for thought.

[pocket-rocket]

Well my electronics experience really is limited. I haven't tinkered enough to do more advanced stuff like program a microprocessor to swing between duty cycles continuously (which I think would be the best approach for fully unmanned cleaning once you set it up). A 555 timer could work once you use something like power transistors to actually drive the injectors, but then you have to sit there and vary it manually like Robert said, by way of a variable resistor. IMHO, the programmable microprocessor is the way to go. It might be a little more complex in the beginning, but once it's set up it will make life peachy keen once it's done being set up the way you want. A 555 timer could work as just a square wave generator driving the power transistors, but the microprocessor can be programmed to do the same thing but be automated as far as varying duty cycles goes. I would help out some how if I could, but as far as my "almighty dollar" goes, my unemployment got cut off ~3-4 weeks ago and I have about $40 to my name, and no income at this point in time. :/ I do wish I could be more help, but I need a job first and foremost right now.

[60dgrzbelow0]

Bob, while i believe all of that is entirely correct, i feel the almighty dollar is making people that know how to fix this particular quirk think twice about opening their mouths... but that's just me, i have a bad habit of expecting the worst from people before i know them. i believe the suggestion a while back of using solid-state relays would work great, then maybe using a 555 circuit to control the duty cycle of the injectors could work good. maybe stick a rheostat in there somewhere and literally have a dial to get exactly what you want to happen... but that's all theory from me. i barely know what a 555 does, let alone how to wire it.

Right as ever.. Robert... But if I had worried about how much money, time and energy I was using while chasing my tail from the jump on this idea... it never would have gotten off the ground. But now...since all the really heavy lifting has been done for all to see...including all the mistakes I made in trying to make it work and what to avoid... now all that remains is for someone who has a passionate idea and a desire to make his (or her...?) mark on something that has a definite need for their creative thinking and the willingness to explain how it works and can be be built...to step forward and show us the way. You already know that I have next to Zero Tooning Knowledge... so in addition to your 555 Circuit idea...Azreal's suggestion about using an aftermarket EFI controller really interests me.... and so.. I will investigate these suggestions until someone comes up with a "Littler Black Box" looking for a friend called "Frank" to lend a helping hand... and worry the Hell out of six to eight Fuel Injectors until they get squeaky clean...inside and out!....

[robertisaar]

Bob, while i believe all of that is entirely correct, i feel the almighty dollar is making people that know how to fix this particluar quirk think twice about opening their mouths... but that's just me, i have a bad habit of expecting the worst from people before i know them. i believe the suggestion a while back of using solid-state relays would work great, then maybe using a 555 circuit to control the duty cycle of the injectors could work good. maybe stick a rheostat in there somewhere and literally have a dial to get exactly what you want to happen... but that's all theory from me. i barely know what a 555 does, let alone how to wire it.

[60dgrzbelow0]

Actually... That notion is pretty clever... but perhaps using some older handheld harware... like a Palm Tungsten 2E? I'm with you completely though... about the idea that there must be SOMEBODY who knows enough about electronics and the electricity used to control EFIs to make SOMETHING very straightforward enough to cycle six or eight of them at a time with variability. I don't know... perhaps our folks don't think I've been genuine enough in stating that everything I've put out there on this unique machine is completely free for the taking and that I have absolutely no interest in either marketing or selling "The FrankInjector Machine" as a going concern... and this prevents anyone talented enough to see how easy it would be for them to make or to even write a diagram that others could follow and simplify the only really complex problem this thing has encountered since the "FIM Gen III" unit proved itself in all other aspects while being developed. There HAS to be somebody with enough ego and know-how to want to show us all what can be done... Yes?

[Tbay99Venture]

I would very much like to try out on " The FrankInjector Machine" and end the problem and complications of having to employ two OTC units to get the job(s) done correctly

That particular problem should be straightforward for a tech or hobbyist to solve. I'm surprised nobody has some forward yet with the range of talent on this forum. The idea that we need one more custom box with a way to flash updates (that you probably have to buy) over a USB makes it one step away from an iPhone app to run Frankenjector.

Or maybe I'm just bitchy today?

[pocket-rocket]

Or maybe design a circuit with a cheep microprocessor that drives some good power transistors to control the fuel injectors and program it to speed up and slow down continuously.

[Azrael]

Maybe you can also find a cheap (cheap might be a stretch), used aftermarket efi controller (Haltech, megasquirt, etc) and program it the way you want it to operate. I'm not sure how well it'd cooperate but it might be an option.

[60dgrzbelow0]

Thanks "Tex"... but as you well know... the usual formula and movie plots is for "The Little Guy" to dream up something their engineers never thought of...and wait for the "Evil Corporations" to steal it from THEM!

Okay... A New Development!

I have to say that I am fairly excited about a recent development concerning an email I received from an otherwise "shy" member/reader of this thread who also just happens to be an Electrical Engineer interested in developing his own "electronic control panel" that would serve to give some greater variability over how to activate and actuate Electronic Fuel Injectors. He seems very keen on what has been done with "The FrankInjector Machine" so far, knowing full well that it is wanting a more practical means of actuating the EFIs during the cleaning process than my solution of using the twin OTC devices...even to the point of allowing this to occur "unattended" (which I would never recommend). I have agreed to test out whatever device(s) he comes up with... with his permission and my understanding that he may want to design, build, market and sell what ever it is he manages to get to to work is a separate project from my own "Open Source" EFI Cleaning Machine. I'll keep everyone "posted" as we correspond about this idea.

Okay... As much as I am seriously interested in this Gentleman's idea about creating his own OTV means to actuate and variably control Electronic Fuel Injectors... I'm a bit frustrated with the idea that he has not addressed this matter openly in this forum thread where he and others would have a better chance of interacting about what he is attempting to do. Instead, he is trying to accomplish this in the vacuum of sending and receiving personal emails to me on the matter. The attached screen prints are of his latest communique, stressing that in all matters relating to my end of this project... and my response to his request.

The man may already be a member here... but since I have no way of knowing or confirming this to be the case... I made my latest ( and last) request for him to join us all here and try to help him with the development of his unique EFI "Driver" device. I will not be interacting on this matter unless it takes place on the site of its original posting: www.60degreev6.com. There are many of us here who might enjoy working with this engineer in making something that I would very much like to try out on " The FrankInjector Machine" and end the problem and complications of having to employ two OTC units to get the job(s) done correctly. It has been a very long time since this thread was started in order to answer this very specific question.. .and it has not been solved yet... But I think this fellow is very capable of doing the job...and I am hoping he will come here and share his ideas with all of us and complete the last and final necessary piece of work.

[pocket-rocket]

Woops... Right... Sorry about the Math Faux Pas... That would be around 153,000 miles @ 18 MPG... Thanks... Still ... it is the same amount of fuel being pounded through those injectors at around 43.5 PSI... Ouch!

I was just teasing there, not really being picky about the math (was more curious as to what kind of mpg that was, honestly).

[60dgrzbelow0]

You driving a dually there, Bob? (I ask because that's just over 11mpg)

Woops... Right... Sorry about the Math Faux Pas, P-R... That would be around 153,000 miles @ 18 MPG... Thanks... Still ... it is the same amount of fuel being pounded through those injectors at around 43.5 PSI... Ouch!