Type 3 D-Jet Harness Thread

[KTPhil]

This thread is intended to be a sort of journal of knowledge about making a new D-Jet harness. Lots of discussion in other threads, pasted here to get us started. But I sense there is enough interest in this topic to warrant a separate thread, which I hope will evolve into a Harness Build Guide, eventually to include wiring templates, parts lists, tool needs, etc. I'll start with some cut/paste:

[rosevillain]

L-jet offers us new style injector connectors, but is there also an upgrade for the double and triple plugs that would be needed for the TPS, pressure switch, and trigger points?

[KTPhil]

OKay, so as far as materials go, can we answer:
1) What gauge wire is needed? For simplicity, if we can use a single gauge that will be helpful, even if it is overkill for some circuits.
2) Source for connectors? There are many, so below is a list I will edit until I get it right. Eventually I'll build a table with sources and part numbers.
3) Sleeving: What type is best? Some may need heat-shrink, others stiff for neatness, others soft so it can be bent easily while installilng. Any recommendations?
4) How about the rubber boots? We need various sizes.

CONNECTORS:
1) ECU card edge connector, plastic shell, 25 pin
2) Injector 2-pin connectors and boots
3) multi-pin for TVS (2/4/5 depending on year)
4) multi-pin for MPS (4/5?)
5) 3-pin for trigger points
6) Ground connectors (either porcupine or those at each injector pair)
7) Head temp sensor connector, 1-pin
8) Crankcase/intake air temp sensor connector (2-pin)--not sure if these are identical?
9) Power conncotrs leading to relay(s)
10) Pump relay connector (in engine bay)
11) Cold start jet connector (2-pin)
12) Thermo-time switch connector (single connector)
13) Ground at MPS on later models?

Any I'm forgetting?

[KTPhil]

Tools:

Ray, can you identify the crimper tool?

Any tools needed to revamp the ECU card edge connector?

Any other tools needed?

[KTPhil]

Another issue to start data collection for is the variation over the years. It may be helpful to create a chronology for T3 harnesses, eventually to include specific wire number changes (is purpose or length or connector type). This will let is condense the variatin into one, or at least a small number of, variants. My hope is that we can make a "common pegboard" layout for actual wire length, then a "connectorization plan" that will tell us how to wire the ECU end, sleeve/bundle the wires, and finally to crimp on the connectors and boots. The more common we make the raw harness, the easier this will be to manage.

So there are at least these variations based on ECU generations:

'69/69 with the pressure switch

'70/'71 with no pressure switch, new MPS, new TVS, new central ground

'72/'73 with new MPS, TVS

When did the crankcase TS become the intake air TS? Did this create an actual wiring change, or just re-routing?

One last complication is for those which have the retrofitted cold start system added, but we'll get to that later.

[KTPhil]

One thought is to make the harness such that we can easily add a fixed resistor to the head TS circuit, or even a pot. This has been a common fix over the years, and we may as well admit it and accomodate this.

So once we leave the stock world, are there any other changes? Is there benefit to adding some test points, either for those withotu the VW testers, or for common, simple checks/fixes so that we don't strain the ECU connector? The danger is that it adds potential for added inductance, resistance, etc.

[Tram]

Phil:
Retrofitted cold start systems can run their own auxiliary harness. All you want is a 5th injector to squirt a shot of fuel while cranking. There's a special terminal on the starter for this.

As for the crimping tool, a German made Knipex wire stripping/ crimping tool has the proper 4mm (IIRC) crimp dimple.

As for the connector at the ECU, I'll have to pull a harness out of my stash and look at the big connector. If memory serves, it has the little "catch spade" that sticks out and lodges in a little notch inside the plastic housing like all the others do. You just stick a thin metal strip down the notch in the plastic holder to release the locking tab, then slide the metal connector out. I need to re- verify this, though.

About six or seven years ago, I was able to get an FI repair kit in a metal carrying case through Wurth that had all the proper ends. I need to get caught up on some projects and then look into connector ends. I know for sure that a lot of D-Jet connector parts are still available through Mercedes- Benz.

Ray is correct in saying that a lot of FI problems are harness- related. Again, in my experience, though, it's been ONLY at the connector ends and the ground ends. I have yet to see a harness just go bad internally unless the engine has been on fire, or somebody tried to forcibly make it go on a route it wasn't made for.

Back in the 80s, I built a harness for a beautiful 1973 I picked up that had been "upgraded" to a Weber Progressive. The PO left most of the FI stuff on the engine and the ECU in the car but cut the harness away. It took me 3 or 4 hours with a wiring diagram and a bunch of connectors I got from a buddy who was a tech at a Volvo store.

Let me tell you a little story. A couple years ago I was trying to help Anniehum out long- distance with an FI problem on Savanna. It sounded like a vacuum issue but it was one of those WTF issues that just wouldn't get resolved. I knew that Annie had been doing LOTS of connector end repairs on the harness and she believed it to be a harness issue. I found an NOS harness in the box for $99 on eBay, picked it up, and loaded it and my diagnostic tools into my Mercedes Bus for the trip up to Washington. (I was also going up for Cruise for the Cure.)

We put Savanna through her paces very late Thurs nite after I arrived, and decided to install the NOS harness Friday AM after the low battery was completely charged.

We installed the harness in about a half- hour. Battery was fully charged. It made no difference.

After some checks, we determined that the MAP was an open vacuum leak. We went to her other property and grabbed the one off her parts car, installed it. It ran better, way better, but not 100%. We tried a couple other MAPS from The DDB that Martin brought with him to the show. Finally got Savanna to run about 85- 90%. Annie said that the engine was wearing and would be due for overhaul one day in the future, anyways.

Fast forward about three months. Annie calls. Her generator "exploded". She installed the one from the parts car as a stopgap while waiting on an alternator conversion, and... guess what? The old Savanna was back. The old generator had had new brushes, etc. but the windings must have been getting weak. In our haste to get the little Square running for CFTC, and then with me needing to get back down to Oregon right after, in our haste we never checked. The problem never showed up in the warning light system, or even with accessories being run. Yet, it was there.

The interesting corollary to all this is that, in order to make the replacement generator fit, Annie had to replace the existing "eye" connector ends with push- ons, as this was a '67 style generator. Could the connectors have been the real issue? About a year previous she and her friend Danielle were trying to diagnose a sudden "shutting down" problem on the car. THAT issue wound up being a defective connector end on the power feed wire from the switch to the coil.

Anybody seeing a pattern?
_________________
Немає виправдання для війни! Я з Україною.

[Bobnotch]

Phil, you might want to make a note somewhere in there that mid to late 71's got a computer diagnostic (CD), and parts of it got added into the FI wiring harness (just like the 72's and 73's). I'm only mentioning it, because the 72 and 73's also had EGR and other functions tied into the FI harness, along with the computer diagnostic sensing a few items. I don't know how much the CD was involved in the 71's, as the two 71's I have here don't have the CD at all, and never came with it (they're both January 71 cars).

I will say this though, you've got a lot of great info in this thread, and you're really just scratching the surface.

On another note, I thought I had read Ray's talk about the cycles of the connectors before, and that's part of the reason I haven't installed the new harness as yet. I've got a 71 t-3 case here that I want to send to Rimco for some machining, then I'll build it up and put it in the 71 Notch, along with the new wiring harness. I'm not real happy with the engine that's currently in it, as I think it's got the wrong cam for FI, as it ran really sweet with the carbs, both the Solex carbs in the 65, and the Webers in the 71 while I was searching for my problem (as posted in the FI sticky). I figure I'll probably just put that engine back into the 65 where I "borrowed" it from.
_________________
Bob 65 Notch S with Sunroof
71 Notch ...aka Krunchy; build pics here;
http://www.thesamba.com/vw/forum/viewtopic.php?t=249390 -been busy working
64 T-34 Ghia...aka Wolfie, under construction... http://www.thesamba.com/vw/forum/viewtopic.php?t=412120

[Russ Wolfe]

I dont think that they ever hooked the FI harness into the computer diagnosis plug.
_________________
Society is like stew. If you don't keep it stirred up, you end up with a lot of scum on the top!--Edward Abbey

Gary: OK. Ima poop.

[KTPhil]

This thread will take a while to answer all the questions. Right now it's more important to bring up all the issues than to solve them fast. Keep 'em coming!

If anyone has a NOS harness, a photo with it laid out like on the motor, and measurements of wire and sleeve lengths when straightened out, would be a great help.

[KTPhil]

[raygreenwood]

Tram is spot on. The wires were not a problem...just the connectors. Also, though they worked the 6/6 nylon connector plugs got to be a problem over the years.
I have been experimenting over the years with making nicely molded plugs of very hard, high temp expoxy....that fit L-jet terminals. These are only needed on the MPS, TPS and trigger points plugs. The injectors and ambient air temp sensor easily take slightly modified L-jet plugs for a very tight fit.
Also the nylon plugs were not really that precise in fit. Some of this was also the molding of injectors and other connectors themselves.

Yes...its true that on a properly tuned car with with an OK harness...you might find that swapping in a new one gets you nothing....visible.

This was also part of the commentary of the AMP documents I found. When in a statci environment...decently tight plugs and connectors seem to have no real issue. But...add vibration ...like from revving and driving and it can selectively have failures in variosu rpm ranges that are invisible at idle. This was found to be common among cars that run great...but seem to have an untracable stumble in certain ranges of rpm when under load.

Knipex makes great tools...but not great enough compared to say...Paladin...to make it worth spending three times as much. I can get a Paladin 3000 series crimper frame at someplace like Frys electronics for about $50. Spend another $40 for two specific sets of dies and you can crimp everything terminal in your fuel injection system...factory correct....and every other terminal in your entire chassis harness.
My crimper was bought just to do EFI harnesses....but has payed for itself many, many times over in work all over the car.

The reason crimping is better...is sheer uniformity....and vibration resistance that solder does not have. Not to mention speed.

Yes....going to one wire gauge will be fine. I noted that there was nothing inherently wronng with the original wires. There is also nothing greatly special and no real worth to saving them. As I think Brad Anders noted and also Jeff Bowlsby found ...there are at least three (IIRC) wire gauge sizes within D-jet harnesses.
I have found that with very modern low resistance wires...there is no reason to maintain using the different gauges. They become negligable.
Also...original harness you will find ...maintained equal wire lengths in certain areas....like left and right injector channel wires from the triggers and the injector wires themselves.
That is also unnecessary with modern quality wires.

Using stranded oxygen free wire keeps noise and potential down low enough that all wires can be run directly to where they need to be without maintaining exact lengths. Good crimping, the use of varnish on crimped ends and proper cleaning, polishing and sealing of male pins on components also make a difference.

Because this system lives or dies by resistance changes...its worth it to be overkill on anything that can reduce resistance or voltage potential during heat expansion, vibration and high moisture environments.
That means I have no qualms about crimping on gold plated ground connectors or experimenting with things like all silver wiring (I'm doing some tests on a few strands now) and other seemingly silly things like that.

You must bear this in mind.....if the ECU is reading the output from a sensor...MPS, AAT CHT, trigger points etc......there will be a +/- tolerance to the sensor itself.....and a +/- tolerance to the wire and connector sets in between that "filter" the result the ECU sees. It has been a lot of these problems that has affected the tunability outside of what is generally seen as the stock abilities of D-jet.
The MPS is highly tunable....and the ECU is capable of reacting to careful tuning levels not seen in stock D-jet vehicles.....but the harness....most especially the connector system....and then coupled with very old harnesses...limits that tunability greatly.

This is not even getting into the fuel pressure stability issues that D-jet had. If you do not have 0.5 psi stability at all rpm ranges on your D-jet system (something the factory never gave it)....you are missing out on a HUGE range of tuning possibilities. Get a feeder pump. It helps.
Just remember....the ECU assumes you have 28 psi. Its what the system is designed for. It has no sensors to tell it when you DONT have 28 psi. Every 1 psi equals 2.8% variance from what is programmed. Thats big...considering stock D-jet cars typically vary by as much as 2-3 psi whenever you rev it.

Recycling the ECU card connectors for D-jet is simple...but tedious. Thankfully there are only 23 of them. I take old ones...and craefully file and dremel away the crimped barrel area so that ther is nothing but a nice unifrom flat "pin" of about 7-8 millimeters long. I use a micrometer to make sure that all in a set are the same width and thickness.

Then.....drop by a local jewelry store that restores watches of old items. Pay about $30-50 bucks to have all 23 of them gold plated to about 8-10 microns. If you get along with people well...you can also find rwheel rim shops that will toss you connectors into the same vat as a set of rims for a low rider for a little less. Just make sure you get them all wired together on a "sprue" with their own + and- connectors on them to make it easy for them. I have just recently gotten a new Rheostat to start doing these little things at home.
Then...take a telecom grade 18-24 gauge butt connector (surprising available at radio shack)...strip off the plastic outer shell...do a few mods with a file....and crimp that onto the flat pin on the card edge connector. Use a very small coaxial cable crimping die on your ractheting crimper set on #3 depth....and you now have a permanent new connector barrel on the end of each ECU card edge connector. These work great.

yeah...you can burn a weekend just working on these 23 connectors. But to have a brand new better than factory harness....is worth it. Ray

[JSMskater]

Ray-- I'm not really following the process for restoring the car edge plug... would I be leaving the original connectors inside or...? I guess I just don't follow what parts need to be filed away and then gold plated?

I had a thought a while back of using one of those big round engine harness connectors used on modern cars, they accept something like 30 connections, so plenty big enough for our purposes, and crimping together a female end to the original wires (cut back of course) and a male on the rest of the harness, so that we'd never have to mess with the ECU plug again... we can just plug it in once and then from then on plug/unplug the larger more modern connector that wont fail after some cycles if we need to remove the engine or what have you.

my only issue with this is whether the extra connections in the plug could lead to resistance issues. I'm not particularly worried about heat moisture, etc, since these big plugs live in the engine compartment on modern cars, and feature a screw on type sleeve with an o-ring to ensure a tight fit that wont vibrate loose.
_________________
71 Squareback-FI -- 73 Bay (subaru powered)
TOOB Member #3
I make D-jet FI connectors

[raygreenwood]

For definitions: The connector is the little metal part crimped onto each wire. The plug is the plastic housing they fit into

It goes like this....with a small flat jewelers screwdriver, push the prong on each contact of the card edge connector plug. Remove all 23 connectors from the plug. With a wire cutter, cut off the wires from each one.

Take a close look at where the original wire was crimped onto each metal forked shaped connector. You see a small round, split area smashed flat around the bundle of wire strands. That is called the "barrel".
It is that barrel we are filing off. Or more precisely...we are just filing away the two "flaps" on the top of the barrel that have been crimped down onto the wire strands. When you file away the little metal "flaps"...you are left with a "tang"...about 8-10mm long...about .030"-.035" thick and about .050" wide.
The area you want gold plated...is actually just the fork end where it contacts the top and bottom side of the edge of the circuit board. But...its no big deal if you plate the whole connector.
The gold plating is just "gravy". It works just as well as stock even if you don't plate them.

What we are doing....since you now have no "barrel" on the end of the connector with which to crimp on new wire....is crimping a new barrel onto the end of the tang of the connector.....with a very tight crimp.

No....if you crimp and seal this properly it makes no resistance issue whatsoever.

I have over the years, tried almost every method of adding a totally new plug to the board of the ECU. I finally actually got a full set of brand new forked card edge connectors from AMP.....as a sample....because they would not sell them to me. They have thousaands in stock, but are not allowed to sell them to the public because they were a proprietary part owned and used by Robert Bosch.
Knowing they are rare as hens teeth....I have not used them yet. Instead I started experimenting with "recycling" the original connectors. I tried drilling out the old wire strands from the barrels. I was successful on about 2 in 10....but it left not enough metal for a really strong crimp and was very hard to get the wire in cleanly.

I destroyed several ECU's...by adding new card edge connectors.

First....its very hard to find card edge connectors that have the correct terminal spacing. In fact...its nearly impossible. Second, the only ones that have the proper spacing and terminal quantity....were solder in. That meant removing the board from the ECU and drilling a hole in each edge circuit path and soldering on the board back. That has LOTS of prtoblems. Then....each wire either uses a molex pin type connector...which toatlly suck and have no place in automotive fuel injection...because they have no tension device built in. Those are worse than D-jet connectors. Or....the other terminal and card edge connector design was the curled spring terminal type....single cantilever. This also has no business in autotive units. Lastly....the type you see on the relay board of Megasquirt.....which is my only beef with them. A solder on card edge connector is not a hideous problem. A card edge connector that has screws that hold each bare wire into it....are absolute JUNK. They have no place in automotive reistance sensitive places. Because....the only way you can ensure that teh screw is tight enough to not come unscrewed when it starts heat cycling...is to tighten it so much you either damage the screw...or damage the wire strands.

Make note....screw tension type card edge connectors like I am describing above...are crap. They have very specific places where they are desired and are excellent. Most notably in static machinery with climate control (computers, printing presses, HVAC equipment..etc.)....but not in a car.
When a cimp is done properly.....it compresses all of the strands of wire into a virtual solid. When a crimp is done properly...there is just enough excessive crimping force that even normal heat expansion cannot cause enough growth to relax the crush force on the wire strands. On a properly crimped terminal....you nust not have enough force to horribly deform...or sever...ANY wire strands. It takes practice, the proper wire to barrel gauge....and an adjustble crimp pressure crimping tool...which are easy to find.

No....with D-jet....because of the design of the edge of the board...and the terminal spacing....and the needs of a heat resistant, vibration safe tensioned connector.....we are better off with what we got.

The terminal recycling method works perfectly. I may start producing and selling sets of them. But..they will not be any cheaper than what new terminals would cost if they were available at a dealer. That would be about $4-5 bucks per terminal. Yep....thats what terminal ends cost at most dealers out of their connector box. Too much labor involved.
I could make people a sample and send one to them. One look at what it is...and you would instantly know what you need to do to produce your own. I can also e-mail pictures of how to do it.

The wires in any D-jet system...now 30+ years old...should be thrown away. I can't think of a single place....with a car as unique as my 412....where its worth breaking down while driving it...and having to walk away from it and leave it alone for any length of time while either hiding from weather, getting out of "the hood" or waiting for a tow truck.
The loss of your investment is not worth the risk of driving around with "sh*t" wires. I would not drive a D-jet car with a factory harness in it.

The people that are the worst about the harness are the 914 guys. Too many whine and moan that they want stock D-jet terminal ends to maintain the historic value. What a load! Then why arn't they driving around on 1974 Michelin XzX's ?
Because driving on antique tires for historic value is not safe. Driving around on original, defective D-jet wiring harness.....which will stop you dead and unfixable in the middle of nowhere if you don't have connectors, crimpers and Voltmeter in your trunk....is not safe either.
Besides that....a great many engines flat out burned up because of improper fuel mixture given by a decent fuel injection system with a sh*tty harness.

When Jeff Bowlsby first started thinking about starting his harness business, he and I spoke on the phone once. He did a first rate job of researching materials, connectors and wires.....but did not (a) have the 20+ years of experience I already had at that point driving D-jet cars and working on them and (b) the information I found from AMP...that confirmed what I already knew from experience about the D-jet harness.

I hear he builds a great harness. But at that time ...we disagreed on one major thing. He was looking for all of the original male/female terminals...(which are junk)....because most of his customers were 914 customers and demanded the original (but defective) connector. Ray

[Tram]

Ray, I'm sorry, but I'm with the 914 guys. I have a really, really, REALLY hard time thinking of the D-Jet harness as "defective from the get- go." It sounds too much like all the redneck VW carb- swappin' "technicians" who have said the exact same thing about the entire D-Jet system for years.

I also have a hard time calling components that have held up 35 to 40 years under way less than optimum conditions "defective from the get- go" But then, I have the same gripes with Jake Raby over hudraulic valves and fuel injection on the Type 4 engine.

To me, taking a known good system and repairing it by cobbling on parts that are not a factory approved upgrade is just wrong. In the last 30 plus years, I've never heard of any sort of retrofit/ recall/ upgrade service bulletins from VW, Volvo, or- perhaps most tellingly- Mercedes- Benz due to known/ surely anticipated problems with these harnesses or connectors. Mercedes used this harness on the 3,5 and 4,5 V-8 engines in their top- of- the- line cars in this era- the W111 280SE 3,5 Coupes and Cabriolets, as well as the W112 300SEL 3,5 and 4,5 series (6,3 had mechanical injection) and the first- generation W116 350/ 450 SE and SEL, as well as the first generation W107 350 and 450 SL cars. Mercedes didn't switch from D-Jet to CIS on these models until 1976.

The evidence that this WAS and continues to be a sound design is just too overwhelming.

And again, the L- Jet harnesses and even into Motronic continued to experience the same issues with grounds and connectors- but, being a simplified system, there were fewer wires.

That's my opinion based on my experience, and I'm sticking to it. Hey- after so many years of this stuff, I can be allowed a little bit of cantankerousness. But, I really think we're making a whole imaginary hurdle out of this "defective straight from the factory" argument.

But, no disrespect to Ray- That's just my opinion, based on what I've seen over the decades. Your mileage may vary.
_________________
Немає виправдання для війни! Я з Україною.

[raygreenwood]

Bear in mind....not only am I factory certified in D-jet (which is useless compared to the experiece I have gained through actually working on it...and logging well over a million miles driving it)......I have worked on thousands of D-jet systems.

Also.....the post mortum research from the company that made the terminal....fits exactly 100% with what I have found. The vast majority of914 owners whine and moan about tuning issues...and are too struck blind by their inability to get past the harness.

How many other vehicles EVER built...used the end terminal design of D-jet? None, Nada..."0"...Zilch...Nyet!
A statement in the test data I found confirmed to me that AMP and Bosch planned to never use these terminals again. R&D into the L-jet terminal began in in 1969. Go figure. Spend millions designing a system datuing to 1965...and then dump it in four years?....does that make sense to you?

No major changes in the terminal design of dual cantilever terminals since 1974....does that sound like D-jet terminals were successful?

Ever gotten 100hp from a basically stock 1.7L (about 90 at the rear wheels)?....with D-jet?...I have. You won't EVER do it with the stock harness.

If you have a hard time wrapping your head around it.....not much I can help you with.

Mercedes had ALL the same problems. I've worked on too many of them. Everyone had problems with D-jet terminals. Mercedes only continued D-jet usage as stocks ran out and L-jet came on line.
By the way....fuel injection systems did not get simpler or with less wires. L-jet has 32-39 major wires in its loom to D-jets 23-25 max. There has never been a simpler system than D-jet. Not even CIS is simpler.
If D-jets connectors were so "well engineered"....as you put it...why not just continue its use into the new systems when they came out in 1974?

D-jet as a system is very sound. The harness is worthless. Ray

[raygreenwood]

Ok...by the way...I was not trying to be glib or snippy in my last reply.
The connector issues to me (and some others)....is a done deal. The existence of the problem has been 100% proven. Not just by me...but by the manufacturers of the connectors long before me ...in the labs...on the dynos and on the road. There are too many good solutions for it to even bother discussing it in my opinion. Beleive what you want. I have already fixed the problem in all of my D-jet equipment years ago....with very excellent results. Do what you want.

There are numerous material deficiencies that are easily cleaned up in D-jet. They made it as precise as they could with materials at hand.....but there were many things they didn't yet know. Its actually my favorite system. I love D-jet!
But....for instance.....in some, not all TPS....there is slack problem at the microswitch of almost 2 degrees that is ONE of the main causes of "the bucking syndrome". It can be repaired but is mainly an issue of build quality of one lot to another across 5 potential part numbers.


You have to bear in mind that the original designs for D-jet came from a Bendix design from loooong ago. The original AMP final blueprint on file for the female D-jet terminal dates to 1966...with the first draft being 1964... .

The D-jet as we know it is basically using an un-updated 1966 terminal design. It eventually used three different materials (copper, brass, phosphor bronze) and two seperate plating processes (tin plate on copper, tin plate on brass, bare copper, bare brass, bare phosphor bronze) in less than six years trying to FIX these problems before AMP and Bosch just decided to design around it with a new terminal (the dual cantileverd L-jet terminal).

D-jet as a system design was pretty solid. The harness was not. Ray

[Tram]

[Tram]

[Tram]