Need some help: FI hard start, dies on idle

[furgo]

That helps me already, thanks! I guess I'll go to the bus to get a feel for the clearance first, and then look at which lenght works best for me. The deep sockets I've found seem to be around two sizes: 50 mm or 80 mm. From what you're saying, the 50 mm would work best, but I'll see if I can fit a 80 mm, which would probably spare me cutting the groove (I don't own a die grinder). Alternatively, if my metal saw blade can cope with it, I might give a go at hand sawing the groove to a 50 mm deep socket.

BTW, whoever else is interested in removing a rusted TSII without decapitation, here's also a good thread discussing this tool and the angle grinder DIY way too.
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)

Decode your M-Plate

[johnsilverlakes]

I had similar issues awhile back on my 76'. It wound up being a vacuum line issue. Something was hooked up wrong. Here's my thread:
https://www.thesamba.com/vw/forum/viewtopic.php?t=641723&highlight=
When you do find the issue, please post how it was resolved!
_________________
1976 Westfalia
Manual Transmission
Federal Emissions

[furgo]

[Hoppy Camper]

Hey there, not to hijack but I bought a new TSii while troubleshooting a very similar hard start. Anyway, installed it and it ran bad. Put the old TSii back in and it ran correctly. I ohm'd it just like your picture (67 degrees F is roughly 20 Celsius, bottom reading of the temp guage, top is outside) and got the attached reading. I don't know much about electronics! How's it look though? Is this a bad TSii?
_________________
1979 Mexico Beige Westy auto
The Wilds of New Hampshire!
www.hoppycampernh.com
https://youtube.com/@hoppycamper

[Tcash]

Forester Chain Saw Wrench 13 x 19 mm
https://www.amazon.com/dp/B01BFVFX9A?psc=1

[furgo]

[furgo]

Continuing where I left it last time, I turned my attention to the TS2. I had covered it with derusting gel previously and all the surface rust is gone, so in theory it should be good to put a wrench on it.

TS2: 1; Furgo: 0

That was the theory indeed.

The extractor socket I fabricated last weekend fitted well, and with my small wrench (the bigger would interfere with the fuel rail) and an extension I managed to get into the tight TS2 space.

But the thing would not budge. I applied quite a lot of torque, but rather than move, I think the bolt walls deformed a bit. There is noticeable play, but they're not round.

I soaked it with penetrating oil once more, and I'll wait until tomorrow to try again. I'll also take a hammer with me this time, and I'll see if light tapping helps.

TBH, I'm not sure if the penetrating oil helps in this case. The crush washer should make the TS2 airtight/oilproof, correct? If that's the case, I'm not sure if it will penetrate to the threads at all.

One thing that I was not quite sure about was something that looks like an additional steel washer in addition to the copper crush washer. I'm assuming it was steel, as it was covered in brown rust. But I might be wrong and it's a bit that protrudes from the case.

In any case, if it's an extra washer, I'm thinking if I could use it to my advantage by removing it (read destroy it) so that the TS2 sits less tightly against the case.

Suggestions?

Spark plugs

After one of hour of tinkering with the TS2 unsuccessfully, I looked at the spark plugs, as ignition was another system I wanted to check.

I removed spark plug #3, which surprisingly unscrewed with too much easy. That fellt certainly well undertorqued, but I'm not sure that's related to my issues anyway.

I looked at it once out and noticed that it's quite black, but the electrode looks quite ok otherwise. I'm not sure if there is too much to read from it: it might mean the engine runs rich, but as I started and stopped the engine a few times during my testing, that might be expected.

The thread looks shiny, not sure if it's oil or fuel.

I don't have my feeler gauge with me as I'm writing, but by eye the gap seems to be slightly wider than the expected 0.7 mm.

They're Bosch W5DC (made in India), which is none of the recommended types for FI. I don't know their mileage, but I'll be replacing them with some the four NGK B5ES I've got in any case.

Update: the recommended Bosch spark plugs from the owner's manual were W8CC. The differences with W5DC are essentially the lower heat code range (8 is hotter than 5) and the spark position (electrode + insulator tip protruding 1 mm vs 3 mm).

Thus essentially I was using a three ranges colder spark plug with a different spark position in the combustion chamber. I think the colder heat range was what was causing the carbon deposits. According to NGK documentation online:

[Tcash]

The crush washer should make the TS2 airtight/oilproof, correct?
If it was air tight it would not be rusted.

Some taps on the TS2 with a punch or dull chisel to break the rust ring will help.
Heat it up with a torch and Melt some candle wax into the joint as well.

Good luck
Tcash

[furgo]

[Wildthings]

Kind of hard to picture, but it is possible to get a nice big dose of spray penetrant onto the threads of the TSII sensor. Measure directly up from the center of the TSII sensor 1 inch and drill a 1/4" hole in the tin. Now using a spray can with a straw on it, spray directly downward through the hole towards the TSII sensor. Repeat the spraying a couple of times over a few days until the sensor decides to yield and unscrew.

If you beat too hard on the TSII with a hammer and chisel you may break the fin the TSII screws into right off, use care.

As far as I know there is just supposed to be one washer under the TSII sensor, it gives something for the sensor to seat against and also holds the tin in place. Use never seize or dielectric grease on the threads of the sensor to prevent corrosion.

[Hoppy Camper]

Thanks furgo, your analysis of my post was spot-on. The ohm's were set at 200K, it was showing 9.5 and I was grounded on the terminal of the tsii. I figured my readings were too high. This is a brand new tsii! Dammit, I'll keep checking it but it did not work when I had it installed and the old one fixed the hiccup.

[busdaddy]

If I was up against a rusted in mess like yours I'd find a deep socket or shortish length of small pipe that just fit over the raised middle of the TS2 where the wire emerges. Then I'd tuck the wire up into the socket (or pipe) and seat it against the flat area between the hex and the raised nipple (where the part number is) and give it a few solid whacks to dislodge the grip that iron oxide has on those threads. If it still put up a fight I'd repeat or weld the pipe to a chisel bit that fit into an air hammer for a vigourous massage. The objective is to crush that copper washer a bit and move the threads into the hole slightly creating a microscopic gap.

If all else fails there's 3 more places to mount that sensor if you F it up beyond use. Some sideways bashes flat against the hexes with a flat drift might help too.
I don't think the big washer was put there by VW, but I have seen more than one installed to keep mangled tin down in place were it belongs, likely not the best way to fix the problem.

_________________
Rust NEVER sleeps and stock never goes out of style.

Please don't PM technical questions, ask your problem in public so everyone can play along. If you think it's too stupid post it here
Stop dead photo links! Post your photos to The Samba Gallery!

Слава Україні!

[furgo]

[Abscate]

I doubt anyone has ECU source code, but has anyone figured out how the injection times varies with TSI reading?

Probably one needs to look at the pulse signal on an oscope to do this.

Does it add a Multiplier to the pulse length, how many steps are there, etc.
_________________
🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🇺🇸 🍊 🍊 🍊

[furgo]

It seems the thread is quite prone to being hijacked, but hey, nevermind. We're all on the same boat

[ROCKOROD71]

The last two posts here I haven't read, its early here, but I DID read your post regarding removing the sensor. Good job making that tool btw, that should work, although the real tool is a deeper socket.
Before I got that tool I literally was spraying that sensor every night and tapping on it with a closed end box wrench and a small hammer. It is close quarters for sure. I removed the spark plug wire and any hoses that were in the way. You can only get small swings of the hammer accomplished there but the goal is not to turn or remove the nut with the hammer but to "shock" the fused rust loose. I would spray the sensor down, wait 15 minutes, and start tapping, intermittently trying to turn the wrench with hand strength, then back to tapping. Maybe 5 minutes of this, then soak the sensor again, and walk away to try another day. Letting it soak overnight helps. Also, with hand pressure, turn it to TIGHTEN it intermittently. Eventually you start to feel a little movement. Even with this movement I was still spraying and tapping, but not getting anywhere. Once I used that tool the sensor came right out with just hand strength (this is a good spot to mention that I am a weak little bitch).

ALSO of note: I too had a steel washer between the copper crush washer and the engine tin, and it was rusted, neigh, ROTTED to hell. I wondered then as now, if this changes the resistance to the sensor? Especially if run of the mill anti-seize and not the nickel stuff was applied to the threads, I assume that thick rust doesn't pass electrons so well. I re-installed a new (used) sensor without the steel washer. If that steel washer is needed please let me know, I can add it back. I haven't had a chance to get back to my FI issue or test drive the bus, with some other car repair issues on my other vehicles taking preference.
ALSO ALSO of note: I didn't try yet but the copper washer on the sensor looks VERY close to a crush washer found in a type 1 oil change kit, the ones that go on the acorn nuts. Someone correct me if I am wrong, but I think that would work, and I have a ton of those....
_________________
1971 STD BEETLE- DD-1st car, 1st love. keepin' it stock! 1600DP, Solex 34-3 Mexi Bosch SVDA Dist NOW w/POINTS
1977 WESTY "KrustyKamper" 2L FI

[Wildthings]

If you are living in a place that uses road salts go to your local hardware store and buy a stainless steel washer to use under the sensor. The sensor receives its heat from the metal of the head via the threads, the washer isn't important to the operation beyond holding the tin in place and giving the sensor something to tighten against.

[airschooled]

I took a small bath in penetrating PB blaster spray this morning. That stuff is amazing for tearing apart a rusted engine in half an hour flat. It gets in the same way moisture and oxygen get in.. Vibration helps speed the process.

As far as what the TS2 does to the fuel map, it richens the mixture by varying pulse width, not timing or frequency, doesn't it?

Robbie
_________________
One-on-one tech help for your vintage Volkswagen:
www.airschooled.com

Road trip reports and tech blog:
https://www.patreon.com/airschooled

[furgo]

[furgo]

Another update, and unfortunately, I cannot report success.

After having removed the surface rust, then nearly a week of soaking it in penetrating oil, trying ice shock rust remover, carefully hitting it with a punch and tightening/untightening with the wrench and the special 13 mm socket... it went like Marie Antoniette.

At some point it started moving on tightening, and also on untightening. I thought it was about ready to go and I could remove it, but essentially it was the cylindrical thread starting to detach from the head... until it broke.

Although it was extremely rusty on the surface (really heavy pitting), I'm not sure if the thread is being held by rust. It might have been overtorqued from the start.

Needless to say, it's been extremely frustrating. Tinkering with the TS2 (which I'm pretty sure wasn't the cause of the FI issue -the engine would die with it shorted to ground-) has set me back one week and I'm now worse than I was before.

On the other hand, at some point in the future the sensor was going to fail of age, so better now near home than in the middle of a trip. Don't despair, focus.

Constraints

As mentioned earlier on the thread, I can't do much where the bus is parked right now. Even towing it might be a challenge. I realize that dropping the engine and redrilling the hole is probably and clearly the best option, but not a possible one at the moment. I'm mentioning this so that it's clear from the start.

With that said, I'm evaluating the other options I have...

Temporary fix

I think, but I'm not certain, that the other 3 possible locations for the sensor are not drilled to accommodate it (on the outer side of each air intake runner, next to their top studs). From the pictures I've seen, the diameter of the holes other than cylinder #3 were smaller. Also, that would require cutting a hole in the tin.

Thus I think I could:

• Simply ground the TS2 input to the ECU. Start will probably be more difficult, but I know the engine started at least once with it grounded. Or,
• Route 2 wires to from the engine compartment to the dash and use a 3k Ohm potentiometer. Regulate it during the start phase and dial it to 0 Ohm once the engine seems to be warm. Or,
• I have a good sensor and I'd like to explore the possibility to use it temporarily, even if not properly attached. There must be a way of attaching and getting it to sense the head temperature by surface contact. I'm open to "creative" workarounds here. Suggestions?

Proper fix, but DIY?

If I can manage to start the engine and it does not die, I can then take it to a shop and ask them to remove the rest of the thread for me.

But I'm also wondering which options I have to do it myself. I've been lucky not to have to use them before, but I could probably use a screw extractor/left hand drill

Pros:
• As the thread is essentially a cylinder that originally contained the thermistor, it's already center-drilled. A left hand drill might work

Cons:
• Space is tight. A power drill can't get in there. Maybe a Dremel?
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)

Decode your M-Plate