injector signal from ECU

[Longbeach412]

I have quickly scanned over this discussion and will toss in a veeeery general 2 cent comment based on my own experience with my 70 D-Jet type3. 80% of all my engine problems ( running rough, rich, cutting, sputtering, bad idle, not running) over the past 8yrs, have been attributed to elect. current supply. Namely, 1) battery, 2) VR, 3) wiring & connections, 4) grounding. Even when the symptoms did not sound or feel electric in nature to me. Make sure these 4 are rock solid.

[raygreenwood]

This is easy to figure out. The fuel pump relay( J-17) gets its power from a single line from relay J-16 .....that would be the the SIMPLE relay that SUPPLIES power...and nothing else...to the fuel injection system. That means, you turn the key and the fuel injection system is supplied with power via j-16. That in turn kicks on the simple timing circuit that buzzes the pump. Thats it.....thats the extent of the control exerted by the ECU. Once the Key is turned to start and power flows....the pump is on its own.
Also...the pump has two simple wires to it. A + and a -. It runs at one speed only...constant...at all times. Where is the control wire to the ECU? There are none. The ECU does not control the fuel pump. It simply has a timing circuit that cuts ito the POWER SUPPLY ONLY....pre-start-up.

Still skeptical? just imply look at the diring diagram....then try this. Pull the plug out of the pump, run a positive and negative straight to the battery....so the pump runs constant...with no harness connections. Turn the key...she starts like normal...and drive your merry way with absolutely no difference whatsoever. This actually saved my tail the one and only time I have ever had one of those relays fail. I drove that way for 1200 miles and two days on I-40. Each time i stopped...I pulled over and pulled the wire loose from groung and the pump stops.
I have never seen a car in my life...fuel injected...wherin the ECU exerted ANY control whatsoever...other than the afor mentioned timing circuit...and most cars do that with a timing relay. What function of the pump... would the ECU be controlling anyway? Ray

[raygreenwood]

The TPS...will never...ever...ever...keep you from starting. This is a very simple part. Toss the book....it tells you nothing . Open up a TPS attached to a D-jet TB...and LOOK. It has a pair of floating contacts that take their cue from any minute movement of the shaft. As you let off the gas the slightest amount, the floating whisker switch...not the wiper arms....but the switch in the plastic fork...pops to the rear. It means that regardles of throttle position ...wether barely cracked or WOTbut above a certain rpm...it cuts off injection...when you release the pedal even a minute amount. Its approximeately 2*. Since they used that early here in USA, then deleted it, about 80% of you will not have that cut-off switch. I think its terminal #17. You can pull it loose...and will have no adverse running affects whatsoever...in fact...in certain tuning situations....it actually helps cancel bucking.
Since you are not even running yet and the starter only operates to 900 rpm....this is not your problem. Yes....you get start pulsation...nothing to do with fuel yet...but power at terminal 18 for both FI pump RELAY signal...and cold start. Check to see if you are getting power at the cold start plug. You should ALWAYS have power there. Its the internal bimetallic element that conpletes they ground. The power is always on once the key is turned...and the starter engages. Like the fuel pump...once above 1200+...its out of the ECU's hands.

Also, from looking at what you have done...I presume...here is something you have not done. Pull the trigger points. DO NOT measure the resistance of the triggers through the wire to the ECU. It does not tell you anything you need to know. Also...it does not tell you if they are even opening and closing...even if they have the same/correct resistance. Also...it does not tell you that when one point closes...that the ground is making contact.
Bear this in mind. If you are reading a complete circuit from both trigger terminals with the trigger plate in the distributor.....then you have a shorted trigger plate...becasue with a 180 dgree cam inside...it is physically impossible to EVER have both sets of points closed.
Next...even if only one trigger point is not making contact......the car will not run. There are two injection channels with a collapsing field system. Once a channel is fired, it must then close....for the other channel to ENABLE. So if one trigger is dirty...the car will fire once for that circuit...firing one pair of injectors....once. Then it will die. You can also read perfect resistance on the triggers...and have them not opening and closing properly due to crusted grease on the cam inside. In short...you MUST MUST MUST...pull the dizzy and the triggers and check them on the bench. There is also the probability that the rubbing blocks are just worn enough on one or both of the points...to cause a trigger not to open enough to break contact.
Also, that last post is correct. 90+% of all D-jet problems are FI harness related. The terminals were defective in design. Thats why they never used them again.....on any FI system. The harness...even when new ...was suspect (AMP/Bosches findings...not mine). Even if you pull every plug and read good wires...just a poor plug fit will keep you from running. Make very sure that the three pairs of ground wires under the plenum have PERFECT contact...or no go.

If your fuel pump buzzes upon turning the key...and it gives the requisite 13-18 psi that little buzz normally gives...then there is nothing in the ECU that is causing this problem. If for some reason the starte is breaking contact to trminal 18 before the requisite rpm is reached...then yes that could cause this.
What is even more common than that...is that you are getting very low fuel pressure just from the bump itself (usually a gummy or old pump or more likely an old fuel pressure regulator).....then couple that with an engine that is below proper state of tune...meaning it wouldn't fire on the first crank anyway....and the fuel pressure runs out before the engine ever exceeds 1200+ rpm...meaning the pump will not stay on.
Again.............wire the pumpo direct to a power source.....and turn the key. Start the car...tune it...then hook the power back up correctly and see what you get. Hot wiring the pump is common technique in ALL fuel injection chapters...as it isolates bad relays...and bad timingcircuits...and weak pump/regulator combinations. Ray

[kh]

Well here is what I meant from Anders sight:

The ECU supplies ground to the FPR to run the fuel pump under three conditions. First, the fuel pump is run for 1.5 seconds after the key is turned to the "on" position and the ECU is powered-up. This pressurizes the fuel system in preparation for starting, but prevents the fuel pump from continuing to pump if the key is left in this position. This reduces fire hazards from leaking, and also wear and possible damage to the fuel pump. A check valve in the fuel pump prevents the system pressure from being relieved through the return circuit after the pump has pressurized the system, preventing fuel vapor formation and improving hot starting. When the key is turned to the "start" position, a signal to the ECU from the ignition switch turns the fuel pump back on while cranking. Lastly, if the engine starts, and the engine speed is greater than 100 rpm, the ECU will keep the fuel pump running after the key is released to the "on" position. If the engine does not start after cranking, the fuel pump stops when the key is returned to the "on" position. If the engine stops running (stalls) while the key is in the "on" position, the fuel pump also stops. This reduces fire hazards in the case of an accident where the fuel system might be damaged.


With key on, and turning a distributor by hand with trig contacts
connected, the pump will start, stop.

thanks for the wealth of info.

I will attempt to characterize the waveform Im seeing.

ps> my injectors that wont fire are all 0 280 150 007
The original set show vw # also 311 906 031b

[raygreenwood]

Hmmmm. I have seen the 007, 009, 005 and even a 003 injector. Trying to place the 007. All of them will work...though the balance may be off because of flow differences or cycle rate ability......but none of that will keep you from starting.

Yes...all of that info from Brads web site is correct. But bear with me here. Everyone makes this system too difficult. Not saying it is simple at all. Just that at this juncture, you are getting way ahead of yourself. Don't even worry about issues such as waveforms coming from the ECU etc. There are a lot of basic mechanical checks to be done before you even turn the key.
There are but 23 wires in the whole system. Not one of them is to be checked with its component attached.
Check them all end to end with the plug pulled from the ECU. They should all read near 0 ohms.
Check the ambient air sensor unplugged at 70 f with the ohmeter. I'm going off memory, but it should be about 3k ohms. It will not obstruct starting.
Check the CHT unplugged. It should be about 2500 ohms. If the connection is poor at the harness, it will keep you from starting.

The MPS should read 96 ohms outer pins and 33 ohms inner pins and must hold vacuum indefinately. If any of those three conditions are not met...it can keep you from starting.

Pull the TB...replace the o-ring...clean the CHT with contact cleaner. Just make sure the whisker switches are all making contact and that each contact patch the accelerator wipers move across..causes the VOM to beep. There are ten per side , twenty total. You can do the basic resistance checks in the book if you want.....but as long as everything moves properly and makes the proper contacts...slightly high resistance is irrelevant to your problem, the switch will function...and will not keep you from starting.

Open the cover on both the FI voltage supply relay (j-16) and fuel pump relay (j-17). Clean the contacts. make absolutely sure of the correct wiring order.....AND.....that the small ground wire that is under the screw on both relays attachment tabs...makes good ground. Those will keep you from starting if upon starter vibration...they make poor contact.

Make sure the terminal at the starter (#18) is tightly crimped and grease and corrosion free.

Don't worry about the cold start valve....it means nothing right now. Unplug it for the moment.

Before you ever get near the ignition key...you MUST test the fuel pump and regulator. First connect a good quality fuel presure gauge to the ring main...1/2 psi accuracy..40-50- psi range. Get 20' of hook-up wire ,two gator clips and two female push on terminals. Plug them into the pump...observe correct polarity it is marked on the pump...connnect them to the battery. The pump will run....fuel will flow. Read the pressure. It should be between 29 and 30 psi. It should be steady. Turn off the pump...check the gauge. It should have dropped to about 16-18 psi...and stay there for at least 30 minutes.

Check all of the power connections to the relays.,..and trace them back to the PROPER positions at the fuse block. It is not uncommon for a previous owner to have diddled with the power connections. You can easily find yourself getting initial power pulse from the ECU to start the pump buzz...then have no power in run mode. Been there.

Now...the trigger points. Testing the circuit by triggering with a spare dizzy...tells you nothing of the one that is in the car. Its usually not the harness circuit that is bad on these...its the triggers or the cam.
The trigger points MUST be removed from the dizzy. They must be cleaned...and MUST have tarnish removed (polish The points)...do NOT file them. Check for play in the cam lobe on the shaft. It is not part of the shaft...and the rivit that locates it can go bad. I am going from memory...but its sounds odd that every other revolution you are getting a signal that triggers that the timing circuit restarts on the fuel pump buzz. It sounds like a bad trigger point plug connection (one of the THE most common problems) of poor grounding or one set of points not operating.

The points should generally read...no more than .9 ohms per side. A maximum pf 1.1 or 1.2 ohms is permissable. If anyone tells you anything else..they are quoting factory crap....and are wrong. They can be as low as .7 ohms. The lower the better. Generally the diffrence fromside to side...should be no more than .2 ohms.

The gap between the rubbing blocks (going off memory here)..should be no more than .930".

You should be able to hook the + from your vom onto one side and the - in the middle....install the points...rotate the dizzy and get a "beep" signal that last approx. half the rotation. Then do the oher side. It should never miss a beat.

Also check your ignition and coil. Make sure that when the key is turned you are not losing power to coil. Check your points and condensor.

Check to make sure you have proper timing. If you are down around "0" degrees advance..there is a high chance it will not start.

I cannot stress the ground bundles under the plenum on the case centerline enough. If they ground, but poorly....you will not run. If the crimps are poor or the wires cracked and frayed...you may not run. These are KEY. Also, that which feeds them, is key. Check the ground strap at the tail cone of the tranny. Check the ground strap at the battery.

If you have done all of these things...just like this..and everything checks out....there is a 99.9999% chance this car will start. Anything after that may very well be the ECU. BUT........you must do ALL of these things....Ray

[MGVWfan]

Let me echo the grounding emphasis. The method VW and Bosch decided upon to ground the FI system to the engine allows resistance to build up in the ground connections with vibration, moisture, and heat cycles (as Ray points out!), so that's where I'd check next. That group of three "faston" (spade, or Lucar in British parlance) connectors under the intake plenum grounds the ECU, the injectors, the throttle switches, and the main power relay. One spade connector grounds two injectors, the second grounds the other two injectors, and the third grounds everything else. If you get a little resistance in one of the injector grounds, it will affect two injectors only, and that sounds kinda like your symptoms to me. The current needed to open an injector is on the order of 2 Amps, so an unwanted resistance of 1 Ohm would cause a drop of 2 Volts, which is enough to cause the injector not to open (I tested my theory on the bench tonight with 12V and a simulated ECU ballast resistance, along with a simulated high-resistance ground). Check the case bolt to make sure it's not loosened up, if it has, tighten it down and make sure the bolt turns a bit on the brass three-terminal male spade connector block as you tighten it (letting the bolt turn a bit breaks through the film of corrosion probably built up on the back of the bolt head). Clean the ground tags with emery cloth, pull on the faston connector while holding the wire to see if the crimp is OK, tighten them up gently, spray contact cleaner on them, and install the connectors, and then check the ground strap tightness as Ray suggests.

The Bosch part number you quoted earlier checks out in the Bosch magic decoder ring reference I have, it shows as the correct injector for all T3 and T4 engines with D-Jet after 1972. The book I have doesn't list VW P/N's for Bosch injection parts, so can't help you there.

BTW...a little analysis of the ECU's fuel pump control circuitry. Bottom line is the ECU requires both trigger contacts to pulse on and off alternately at a rate equal to about 100 rpm to turn the fuel pump relay "on", or have +12V on the "start" input to the ECU, either one. If the trigger contacts don't work right, not only will you not get injection pulses, you won't turn "on" the fuel pump while running, either.

The fuel pump relay is grounded through a circuit in the ECU with 2 transistors and 4 diodes. The relay is grounded by the bigger of the two transistors, and +12V from the starter turns this transistor "on" when you turn the key to "start". The second transistor feeds the first for power-on and running. When you turn the key to "ignition", the ECU receives +12V main power, and a capacitor connected to the second transistor starts charging up. While it's charging, the second transistor turns the first "on", which grounds the relay and turns the pump "on". When the capacitor is charged to a certain voltage, the second transistor turns the driver transistor "off", and stops the pump. That's how you get the power-on "burp" of pump activity. The circuitry in the ECU that conditions the distributor trigger contacts also feeds the second transistor. When the trigger contacts open and close one after the other, the second transistor receives enough signal to turn the driver transistor "on", thus allowing the pump to stay "on" while running. NOTE! For the input conditioning circuit to sense trigger contact activity, one contact has to open, then close, then the other has to open, then close, one contact alone can't make things work by itself. Also note that if the relay driver transistor dies, no relay ground (and no fuel pump) for either starting or running, and you have to use Ray's bypass harness (a good idea to have stuffed in the glove box just in case...). the ECU looks very well designed to me, with good quality components, the only thing it lacks is conformal coating, so don't get it anywhere near water! The edge connector at the ECU, now that's a different story. Mine already has corrosion on the silver (not gold) plated contacts on the harness end, so I can see I'll be cleaning and testing before reinstalling. That silver plating's part of the problem, in areas with sulphur and moisture in the air (like Houston), the silver forms a sulphate corrosion film. I've seen the same thing in some old military comm gear.

[vwbill]

Hey Guys, all I can say is "Thanks" "Really, Thanks!" I cant believe how much I learned is such a topic! Not only have you helped me with my 412 but when I go to my brother's to work on his 70 D-jet T3 wagen! I will be making a fuel pump bypass for my glove box and my brothers! thx again, Bill

[raygreenwood]

MGVWFAN....thanks for explaining in technical terms the simple timing switch that allows the pump to stay on. Its "control" level of the fuel pump is brief...and not that complex, as I was noting. Yes..the grounds are KEY ,KEY, KEY. Also...there are , as noted an even half dozen very simple problems that can happen with the triggers that can stop this thing from running. Nice description of the field collaps that must happen to cycle from one injector channel to the other.

Connectors.....I had my ECU connectors gold plated. That is not silver...it is tin. These connectors are phosphor bronze with a tin coating. Gold plate makes a world of difference, but there is an almost as good way even without plating. Go to a circuit shop...or lectronics shop. They should have one of several brands of award winning contact cleaning pens. Some of these are designed to eat corrosion on various types of contact metal...and then be plugged in "wet". They dry, causing a tight, sealed, much higher transfering surface condition. I will post the names/brands later today.

Now that we are speaking of these particular connectors...I just thought that I would mention...don't ever throw any away. yes they can be gotten if you have an account with AMP....but not for very long , as they had only 1,000 or so in stock in 2001....none to be made ever again as far as they say....also since they are proprietary use parts of Robert Bosch...they are not happy campers about selling them to anyone that is not a Bosch Jobber/licensed vendor. They don't care that the original use for this part was abandoned in 1975. Is just $1,000 of long ago paid for stock that will eventually be culled and scrapped. These are part # 583362-2 in the AMP/TYCO web site. Yes...I have new ones...but enough for maybe two harnesses.
But...I have a method of recycling them. Yes it entails recrimping and replating. No it is not prying the old crimps apart. These get new crimps, via a custom ferrule, that are much more solid than factory...and crimped much tighter. It is the use of the ferrule that dictates they get plating for protection and use of two differnt base metals. Let me know what the level of need is for them. I have not actually priced what theye would be able to sell for. When I do them, I would do 100 at a time. The plating will be the part that brings the price up. 18 micron gld plate is not expensive. The act of it is . These terminals run about $3-4 each from AMP/ Ray

[MGVWfan]

Tin??? No one in their right mind would tin-plate a phosphor-bronze connector clip for use in an automobile, tin just doesn't last in the real world. It oxidizes and sulphates within months of exposure to the atmosphere. No wonder the ECU harness connector clips looks so nasty. I'd assumed they were just tarnished silver, since I'd seen lots of German gear with silver plating in the past. Siemens, Telefunken, etc. seem happier with silver plating than US companies, who seem to use gold plating for high-rel military gear mostly.

Actually, I'd crimp and solder everything instead of crimping alone. When you've got mechanical connection via crimp, and electrical via solder, nothing can cause the joint to fail. Even a perfectly crimped joint can eventually fail from moisture intrusion, or differential expansion.

[MGVWfan]

Oh, yeah, I forgot to mention this. If you have to replace the fuel pump relay, the replacement must have a spike suppressed coil. The OE relays did it with a 470-1000 Ohm resistor across the coil, nowadays it's done with a silicon diode connected across the coil in opposition to the normal voltage polarity. Either get the OE relay, or one you know has reverse voltage spike protection built-in. If you replace the fuel pump relay with a standard ice cube relay, eventually the high reverse voltage generated when the relay coil turns off will destroy the relay driver transistor in the ECU.

[raygreenwood]

MGVWFan....while you are correct about the toughness of crimped and soldered connections....do not solder anything in the FI system. There is a load of research done by AMP and Bosch available in the working drawing archives I came a across a few years ago. Soldered connections do not survive in automotive....and most especially fuel injection environments. They degrade, are to vibration and moistire sensitive and if you find a solder that is free enough of flux and impurities....it will require way too high of a heat level to melt. Also, unless these connectors are machine wave soldered, no two solder joints wll be alike. One of the most common problems is breakdown f the sodered joints in the trigger cotact plate...it creates high resistance...and in the soldered contacts indside of the MPS plug. Trust me...on these parts there is a better way. After grinding the old ferrule off and creating a uniform stub/pin to connect to, I then connect ferrules of high purity copper made for telecom use. Then I crimp with about twice the force of factory. Since the ferrules wall section is thicker than what factory uses...the joint is superb. I then plate over the outside. This includes a new strain relif section too. I Then heat lightly for dehumidifying purposes...ans seal with acid free varnish. Then.....slip heat shrink tube on all that is exiting the plug body...and do not shrink thyem until the pigtail is installed in the ecu...so the plug is now perfectly shaped. Ray

[ubercrap]

Yes, this is been an incredibly informative thread! I'm glad I'm not the only one that needs to be straightened out... Ray, thanks for the blow-by-blow coverage of the mechanical things to check. My back hurts from leaning over to stick my head in the engine bay. I'm finding it difficult to reach a lot of stuff under there, like the grounds at the top of the case and the D-Jet trigger points. As described, a lot of stuff will have to come out. Hell, I've been thinking maybe I should just pull the engine since it probably needs at least a refresh anyway? I'm contemplating selling/junking all my cars except for my 412's, and buying a full-size truck for towing/daily driver duty, thus eliminating car payments and full-coverage insurance (yes, yes, I'm still paying for the GTI), and freeing up money for 412 projects!

[MGVWfan]

Good luck on the FI...let us know how it's going, you've got me for one curious how the story ends!