Troublesome Digifant II 2H engine

Long time reader and Vw enthusiast, but first time poster. Bare with me as English is not my native tongue  and the background quite complex and long. I''ll try to condense the information but it will still be quite a read. Although it would be easier to list what I haven't checked or replaced, I will now go through everything I can recall what 've done to resolve the issue. I'm sure I might have overlooked something obvious and simple.

Symptoms; the engine tends to jerk especially at low loads, for example when cruising a parking lot, but the phenomenon is still there in higher speeds/loads but less pronounced. The jerking occurs when the engine alters between load and engine breaking, hence the less pronounced jerking at highway speeds as the engine tends to be under load constantly.

Other symptoms; the engine has stalled completely a couple of times in very hot conditions, for example in a long highway cue a hot 30-degree day. The oil temp in the MFA reads excess of 116 degrees an the engine runs very poorly indeed. After shutdown the engine is hard to restart and high revs are required to keep it running. After cooldown (less than 110 degrees) the engine works OK again. When parking after driving in warm conditions the car occationally leaks coolant, I assume this comes from the coolant level sender in the reservoir. Hot starts are slower than cold starts, but it idles slightly off in both conditions. The engine tends to give off a slight "fog/mist" when warm only noticeable in the right light, and it is hard to locate, but is coming from the back of the engine (exhaust/manifold region). The temp gauge doesn't seem to go all the way up to the middle of the scale in a normal run, but eventually gets there when the engine is really hot. The fan works as it should.

Measures: Lots of parts have beeen changed or checked during the two years I've owed the vehicle, but the problem persists. At first it sounds like a vacuum leak, but so far I have found none and I have really been looking. The manifold have been apart, the injectors and the throttle body have all been cleaned and gaskets are new. Next culprit was the fuelpumps of which both are new together with fuel filter and hoses. The thermostat has been changed together with both the temp senders. The ISV has been checked and works. The lambda probe has been changed twice, since I didn't trust the universal aftermarket one, now there's a new FAE sender. The earth connections have all been apart and throughly cleaned, an extra cord has been added between the ECU and the engine block (via the ignition coil). The air filter is new, as well as the vacuum hoses. The timing's been set and all of the ignition parts, coil, cords, cap, rotor, plugs have been changed to new bosch/beru parts. The ignition lock switch have been changed because of another reason, but is also new. A compression test has been performed with all values within only slight variations. I've also done a bunch of electrical measuring to find any deviation or oddity.

So far the only main items left regarding the engine management are the fuel pressure regulator and the air flow meter and then I'm out of ideas. Both quite expensive parts. It might just be something simple I overlooked, for instance, I haven't run a performance test of the injectors, I haven't discovered an exhaust leak. New engine mounts is on the shelf to possibly defeat some of the engine's jerking behavior, but it seems to be more than just a mechanical issue.

For those of you that found the energy to read all this, I salute you. 

Kind Regards from Sweden!

Hi and welcome,

Sounds like you are having some grief there.

The 2H Digi engine was only fitted to a few Rivages very late in the UK, so there might not be much experience in this group.

Lets see who responds.

Hej och välkommen.

I have a 2H, but don't don't know much about it as I've never had any problems.

Luckily Briano1234 has one and is very knowledgeable, so hopefully he will be able to help out.

resetting the ecm after changing cooling temp. sender
Common issues that indicate a failed engine coolant temperature sensor are:

Vehicle idles poorly
Engine sputters, might stall
Higher than normal fuel consumption
The part number for this sensor is '025 906 041 A' (always check with your Volkswagen dealer for the most updated part number). The resistance of this unit is approximately 3.2 Kohm at 10 degrees C. If it measures open circuit this will explain erratic idle and throttle speeds especially when the engine is cold.

When replacing this sensor, it is important to also replace the clip that holds it in position ('032 121 142') and the O-ring ('N 903 168 02').

Once the new sensor has been installed, start the engine and disconnect the blue coolant temperature sensor. Rev the engine through 3,000 rpm three times, each time allowing the throttle to close completely. This clears the Digifant ECM fault memory.

Thank you for your replies! I turned to this forum since I usually end up here when googling any way. Tried to find information from Swedish forums, but I'm grasping for every piece of experience/knowledge there's to be found!

Funny you should mention the temp sender as I had an idea of changing it again. A new one  is in the trunk.

But just to point out, this has been changed once already, as a matter of fact, it was one of the first items to be replaced. Together with the black one And I've measured the resistance, but only when warm, and it seems to follow the graph in the Bentley just fine, a tad low, but OK at 210ohms if I remember correctly.

The main reason why I bought another one is a weak suspicion that it could be a matter of quality between brands.  For instance, the black sender reads even lower at 110-130ohms, still the gauge needle is not spot on and thends to vary with conditions. I haven't found any reference values for the black one though. Maybe it's ok and the fault lies elsewhere?

With the new (blue) sender from Febi Bilstein, comes a new o-ring as well as a new clip. I'll be sure to measure the resistance of the both senders as the multimeter now have become standard equipment in my car.

I'm well aware that the ECU needs resetting, but its always good with reminders.

Been busy lately, but I'll do this after work tomorrow and report back. If it really is that easy I'll feel really foolish since I've already tried it once,.

the_whip said

Hej och välkommen.

I have a 2H, but don't don't know much about it as I've never had any problems.

Luckily Briano1234 has one and is very knowledgeable, so hopefully he will be able to help out.

naw what do I know about digi…. . I will read the post in a few…Only owned 3 of them for the last 15 years or so.

Ok, a lot of verbage to go through, but I will give you a quick impression of what I suspect.

The Blue CTS sensor is for the Digifant ECU, the Black one is coolant temp.

The blue coolant sensor can cause cold start issue, cruddy running issues and hot start issues and that is a fact.  

On hot starts unplug the thing and see if the car starts easier… On cold starts unplug it and see if the car starts easier.. Simp effective tests for a iffy part.

I Strongly urge you to read my thread in the archive section on HOW I HATE PLASTIC PARTS….

I still hate plastic parts, but thanks Tolusina. | VW Vortex - Volkswagen Forum

For longevity I will urge you to replace the Plastic water outlet flange with a Metal one, and replace the 2 senders with the 16V CTS sensor and the black single pin sender, in the 8 years I have been using the 16V sensor on the metal flange I haven't replace the flange or the sender. The 16V sender is the same electrically as the blue one and the grounding point to the engine is more effective and less likely to be a iffy grounding issue that throws the signal line off.

Replace the following plastic parts.







found this on Ali,

https://www.aliexpress.com/i/32975280235.html

or this which replaces the plastic ones.

https://www.aliexpress.com/i/33046471517.html?spm=2114.12057483.0.0.f3cb3723Dr3CgA

I know chinese, but hey, it's metal and won't be prone to breakage that the cheap plastic one do, and when you have been stranded by a Broken one you replace it with metal, as it is the weak link in the upper radiator hose connections and subject to heat.  Going back be sure to use copper grease on the bolts, and only torque them to 87 in/lbs with a new 0-ring.  

Ok now back to our thread.

Verify that your Fuel pressure Regulator is good by doing the FPR residual pressure tests.  Why because this will cause a car not to start after a run, because it is allowing all the fuel to by-pass the injector rail and flow right straight back to the tank.  When it cools off it will reset and the car will start fine… I had one that the thing would start acting the BADARSE after a short run or 30 to 40 minutes, then you come out to start the car and it wouldn't start…but by the time my daughter called me to tell me, I would get there, and presto it fired right up…

The residual pressure tests nailed it as bad..once I got my posterior off my head, and looked for the thing.
Read the Bentley or Haynes and just DO IT.

The other test for the FPR being bad is after a run pull the vacuum hose off of it and if there is fuel in the vacuum line as in wet, the FPR is bad.

Vacuum leaks are the Bane of the VW where I have seen them to cause issues are the Air intake tube becomes weak and holed where the ISV connects to the tube, and where the PCV at the top of the Valve cover feeds in to it.  I have fixes for both and a new hose is PRICEY if still available…

Statically Time your engine.  It doesn't require 2 people, starting the engine allowing it to get hot then unplugging the CTS and reving the engine above 3000k to reset the ECU then hold it at 2250 while you set it to 6TDC and get your hands burned by the radiator heat when the fan cycles, and or your knuckles knicked by the running fan.

I only set my timing by using the following link, the originator of the link and I as well as a True Rocket Scientist (She worked for NASA) verified that this worked every time, and cross check it with a timing light was always spot on.

Statically set your timing to the marks. Then follow the below.

https://www.reflectionsandshadows.com/cabby/static-digi.html

I think one of the above will fix your issue.

Hope this verbiage assist you in getting your cabby running greatly.

Thank you for taking your time, and I understand there's a lot to read, but this story have become quite a long one over the years and I wanted you to have as much background information as possible. Please bare with me  and my spelling as I don't have an English dictionary in this web browser.

I'm well aware of the two different coolant temp senders and their function. The reason to why I changed both initially was that the gauge also reads inconsistently, but that's a minor problem if I can get the running issue fixed. The  thought has crossed my mind, that if the black sender gives faulty readings, why wouldn't the blue one also do the same? I have no problems with leakage at the flange, but I'll read your posts more thoroughly later on if it feels the next logical step to take. The thing about better ground for the sender(s?) is appealing though.

Static timing is set, but even though I had to rotate the distributor slightly, it didn't change anything.

Vacuum leaks have been put out of the picture through plugging all accessory hoses on the throttle body (even the brake assist), but to no avail. The rest of the "nessesary" connections have been thoroughly examined several times with the loose hose in my hand almost turned inside out. As a first measure I tried the starter spray even though I dont really believe in that method. I've thought about plugging the ends and submerging the hose into water just to be sure, but have not executed that idea yet.

Anyway, today I changed the CTS once again, since I had a new (slightly more expensive) one laying around as a measure of desperation when nothing else has helped. I measarued the resistance of both the new and the old sender at cold (around 20 degrees today) and I don't recall the exact numbers, but it was something like 2,17kOhms for the old one and 2,77kOhms for the new one. I changed the sender and reset the ECU.

The car drives smoother but not quite right. It also starts easier at warm, and it's promising indeed. And as I sat behind the wheel, I thought that this must be something to do with my problems, since it is significantly better. Later I measured the resistance at warm engine, and much to my surprise it was even lower than the old one, at around 180 ohms. Too low if I remember Bentley correctly. I took the opportunity to reset the ECU again, in case I failed before or if it's not possible at start up of a cold engine.

As I said, it is still quite jerky at parking lots and driving in first gear, then it decreases with increasing load and speed. It feels like I've got a slightly better bottom end now, and over all power has increased. But as I said, still has it's erratic behaviour at alternating accelerating and engine breaking.

One thing I've noticed when doing the static timing, is that the distributor shaft seems to have a significant play in the engine block. Is this familiar as a known source of error, unheard of or normal? I can't recall this being the case with other engines I've handled, but it's been a while since last time now.

Thank you for your time and patience. I really enjoy my cabby and have lots of plans for it once I get it running alright.

Set the static timing, then use the DVOM Method to set it for 6tbdc, it works and there is no guessing, yes there is a little play in the diz.

Do not use the pulleys to set the DIZ.
Set the marks as such.

Crank at 0tdc on the transmission.
Cam will be at 0|T on the front and the dimple at the rear will be even to the valve cover tin, not re-bar.  The Diz will be in the middle of the hash stamped on the side frame or the middle of the plastic dimple on the shield if you don't want to remove the shield and rotor to accurately set it.

Then remove the plug off the diz, to remove the rubber boot, and use the brown lead for ground and the green white for the signal then follow the guide to see the rise and fall from 0-11V (key on in the run not start position).

Resetting the ECU is moot as long as the CTS (blue ) is connected it really won't as you have to rev it over 3000 rpm 3 times.

Then I would check for the residual pressure from the Fuel Pressure Regulator.

LAST BUT NOT LEAST…. If you haven't replaced your grounds from Battery to frame and then frame to engine DO IT… That new ground will resolve a bunch of intermittent issues electrically.


The coolant temp to gauge is a totally different circuit than the Blue CTS, and they can read different…specially if your grounds are iffy, as the CTS is in the engine compartment to ECU and the Coolant temp is the Engine to the Gauge in the console and the 10V stabilizer.  IF you wonder you can a 9v dc battery to test the circuit from the connector to the gauge with no key needed.  Connect the 9v Battery to the sender connector as brown is ground and the colored is the positive.  With that connected your gauge should deflect 3/4 to full hot as that tests the entire system if it doesn't then you are looking at a iffy ground to pin2 of the cluster.  I have fixes for both in the archive section how-to's as list of things for a 90ish cabriolet.




 

OK.

I've set the timing through listening to the activation of the fuel pump, but if I understood you correctly, you think it's better to measure the voltage across the pins of the hall sender plug at the distributor? This is while having the engine fixed at 0tdc and only rotating the distributor? Is that correct?

The ground connections throughout the entire engine compartment has been dissconnected and thoroughly cleaned as well s the grounding points in the chassi of the car. And yes, this includes the thick wire from the battery, to the chassis and engine block. I don't suspect any bad ground at the points I've cleaned. Although I'm a bit concerned where exactly the ECU takes it's ground. Is it thorugh the chassi of the ECU or elsewhere?

The residual fuel test is not yet performed, any link or description of it so I know how to go about? Upon inspecting the FPR I've seen no leakages or other deviations that might cause concern. The hot start issue seems to have been alleviated since changing the CTS. The engine has always started up straight away at cold. I will return with further info as I drive the car more.

Otherwise I feel I've got the senders function straight. Good tip of how to check the temp gauge circuit. Along the description you gave, I recon if I have a 10V source it should read "hot" if working correctly?

I wonder how sensitive these systems are for exhaust leaks? Normally the lambda values will be off if there's a leak before the sensor. But since the probe is so close to the head, I'm not sure how common this would be in this case. The idea here is that there could be an exhaust leak at the head, barely noticable over the "sport exhaust" fitted to the car? Upon decelerating a slight "bluddering" sound can be heard. It might as well be quite normal, the exhasut is after all modified and lacks one of the silencers. I'll try to look out for soot deposits.

Kind regards /Kee

You have to set the DigiFant by the green/white wire to get the proper 0/11v switch, yes the fuel pump will fire but you can be on the wrong end of the drop as in more 0 or more voltage.  Only using a DVOM will be exact.  This is very similar to try tuning bike Carbs or the old twin SU's of the Old Austin/MG/Rovers that I owned in my youth by ear, very hard to do, and well with out the uni-syn gauge you can never get them balanced.

The main ground for the ECU is a Wire flat Braided Cable from the Frame to the securing plate, if that isn't clean and good then you can have a floating ground to it.  The ECU and ICM are mounted to a plate on the scuttle tray that is held in place by 2 screws or bolts in nylon keepers that Isolate it as well as a nylon lock at the bottom.  If you have a voltage difference between the negative battery post and the mounting plate you need to replace the grounds to it or to the frame.

I have seen over the years that the main grounds that VW used were a unshielded Stranded braid of cable that goes from the Battery to the frame, then over to the engine.  I strongly urge you to verify that there is 0 voltage loss between the battery negative post to the positive at the frame point, and the engine as well as to a different point on the frame.  if you have ,05V difference then you are good any more and you are asking for intermittent issues.  lets face it your OEM ground wire is over 30 years old, and has been subject to tarnish build up between the strands, that tarnish can grow and become instead of a single solid braid of copper to individual strands that can't pull the Current effectively.  I have had so many ground issues on different things as in Aircraft harnesses, Cars and large computer server farms that I know what kind of flakiness they cause and is why I usually on autos replace them out of hand, you can use a new oem type of cable or two off the self units.

Remove your battery cables, from the car, then measure the voltage between the posts.  Connect your ground cable and measure the voltages from the positive to the Frame connect, then from the positive to the engine, and go so far as to the ECU plate.  If there is more than a .05 voltage drop then expect intermittents.

Ground effects on your digi, and a lot of good pictures.
https://forums.vwvortex.com/showthread.php?5284503-Ground-effects-and-the-repair-of-your-car

Improving cluster grounds and your gauges.
https://www.volkswagenownersclub.com/threads/improving-your-cluster-connections-and-ground.24949/

Parasitic Drain video.
How To Perform a Parasitic Draw Test - EricTheCarGuy - YouTube

Yes I am anal on good ground as to answer this riddle.

What does a good cup of Coffee,  Cars, Computers and Divorces all have in common?   They all start with goof grounds.

Manifold leaks are quite common to the oem 4-1 manifold to the sport down pipe.  So much so that on the 3 that I have owned all had cracks some so bad that I couldn't pass emissions tests over here, one whistled like the old beetle… So I replaced them one with a pace setter header, and the last 2 with the g60 4-2 manifold and tt-tuning dual down pipe.







A very poor design, with bad motor mounts it will crack eventually they all do… On the tt-dual down pipe I had to weld a bung for the o2 sensor on the lower right side of the down pipe, as well as on my Pacesetter header.

But the difference in the engine sound and seat of the pants hp gain was magnificent at least 35 ponies.  

I was originally going to install a Cat delete header on my 93 once it got out of emissions testing, but never did.  So that Stainless Steel Beastie is sitting on a shelf.

Don't discount the CTS ever, it is a Cheap to replace part that can cause so many issues in different ways, that I used to carry a spare in the glove box.  On my 92 when I drove 400 miles to pick it up, I got another 500 knocked off because of a hot start issue that was fixed by removing the CTS sender harness.  Little did I know that in 100 miles not only did I have to replace the CTS, but a bad lift pump before I got it home.  Don't get me started on what happened 3 weeks in to new ownership and the previous owners repair person.

Residual fuel pressure tests are outlined in the Haynes and the Bentley, they are easy to do with a gauge that you affix to the Fuel Rail port on the passenger side of the fuel rail opposite to the FPR.  It measures pressure drop over minutes…

Section 7 sub class 6 in the Bentley pages 167-169







I have had a couple go bad over the years and it was a couple of different ways that the residual tests nailed.

if you haven't gotten a Haynes or Bentley yet you might want to think on it….

Thank you for all your great advice. I will double check my grounds and the timing according to your instructions. I have a Bentley already and it's well read already, but it's good to know what chapters to check out.

The engine has been running OK for the last couple of weeks, but with too little spare time, too many projects and only a few weeks left of cabriolet weather, I won't initiate a bigger tear down at the moment.

But I will try to evaluate the different apporaches in this matter. And get to it eventually.

Just wanted to share an update with whoever it might concern.

This year, the search for a better running car has continued. i will probably forget one or two details but here are some of the more major things.

It started with the MAF, where I dismantled the original to try to clean the carbon traces but with no difference made. I also got my hands on another one but swapping them changed nothing either.

The fan has always been very noisy, this together with something askew with the engine's temp, made me try to get to it again, this is despite the temp gauge in the dash reads low. The engine "feels" very hot.

So in doing so I also changed the radiator and discovered that the old one had a minor leak (which hasn't affected the coolant level particularly). Whilst at it I also changed the thermostat (for the second time). This lead to no major changes except for a quieter fan and a more tidy engine bay with a shiny rad and painted fan shroud.

I suspected I might have missed something in search for vacuum leaks so a dismantled the throttle body and let it sit in the ultrasonic cleaner for a couple of hours. It was surprisingly sooty considering that I've cleaned it a couple of years ago. The idle screw's o-ring looked fine and none of the hoses had any apparent cracks.

With the TB off I performed a compression test (again) on the engine and could not find any alarming deviation, although the values differed a little. I preformed the test three times to make sure it was consistent.

In my strive for a nicer looking dash I also installed an oil pressure gauge together with a new sender. This however gave me some new information on the engine. The warning light went off (later confirming threshold of 0,5 bars, original is 0,15 bars?)  and the needle barely above the zero mark. With an oil change to 20W-50 this improved, but it still sinks below 0,5 when warm at idle.

I'm not overly concerned riding this thing in normal traffic as it keeps around 2-3bars, but in long traffic jams on hot days I feel a little bit uncomfortable.

I've also changed the main ground strap from battery to body to engine for a brand new one but it made no difference at all.

Three of the four engine mounts was replaced, i still have the one on the timing belt side to do, but the three i changed was shot so that felt good. That affected the ride in less jerkiness because the engine sits more firmly, but the cause for the jerks is still not tracked down.

I'm still down on new ideas, check the timing once again, and perhaps perform that fuel pressure test, but otherwise I cant track it down. Since discovering the low oil pressure issue, I have one more reason to go the 2l engine swap route, but I'm very doubtful I wanna keep the digifant. I might find a crack in the exhaust manifold or accidentally fix a bad ground in the swap, but as you hopefully can relate, it feels like after market ECU or even carbs is the way to go…

Thank you for your time!

Don't discount the Digi.  Some knock it, I don't.

I guess I am a little perplexed by the term "Jerkiness".

Vacuum leaks can cause a plethora of things as in poor idle, fast idle but I never have had jerkiness as a result.

The timing can be set with a DVOM no need for a timing light 2 people and holding the rpm at 2250, while the CTS is unplugged.  The CTS sensor can also cause poor idle, fast idle, and well no idle.  It is usually the first thing I usually replace out of hand, after I test it and the new replacement.

The only thing to look out of is to remove the Hall Sender plug off the Diz, before you try to remove the boot to get to the green/white wire. then reconnect it.

The Key only needs to be in the run position, the engine doesn't have to be up to temp, with the fan cycling one time, nor do you have to remove the CTS or hold the thing at idle.  All you have to do is to follow the how-to.

Statically setting the Timing with just a silly DVOM.
https://www.reflectionsandshadows.com/cabby/static-digi.html

I have had vacuum leaks in the HVAC lines that effected the running as well, which is why I usually replace all the cloth covered bits.

You mention grounds, and those are the biggest thing I know of that can cause a whole lot of grief.

Here is a how-to and if you scroll it you will find pictures of most of the engine bay grounds.

Ground effects and the repair of your car. | VW Vortex - Volkswagen Forum

Knowing if it is Manual or Automatic I suspect is another Jerky issue as well Autos can be jerky if the fluid is low, or the ATF pump is weak.

Most poor running on the Digi that I have found is related to bad wiring (HV and Plugs) Grounds, or fuel delivery, and or vacuum leaks.

I did find that running NGK Plugs instead of Bosch really made a difference.

Oil pressure.  The only way to accurately determine is to use a mechanical gauge.  Where VW placed the Sender is at the bottom end of the pressure trail.  The OIL pressure is considered low if the oil is at 80C, and the idle is at 2k then you should have 1.5bar.  

If you bought a oil pressure sender off of ebay, and it cost less than 50 bucks it is probably a Universal, and too low.

The oil pressure sender can sits on the side of the head towards the tranny.  one side of it is the low pressure, the other is for the pressure gauge.

The High pressure when running is a White Sender that is on the Oil filter flange, it will light the light and well sound the buzzer.  

A flickering light at idle is usually the sender on the head.  

You can remove the extra boss on the oil filter flange and re-locate the pressure/low can switch to the oil filter flange.

You can alos upgrade the oil pump from the 32mm gears to the 2.0L 36mm gears, which increases the Oil pressures, and delivery.  You will need to use the pickup tube out of your existing pump as the 2.0L pick-up is longer and you can't get your oil sump back on.

My 2 pence.

Digi''s I can understand, it is CIS that drives me batty.

Briano1234 said

Statically setting the Timing with just a silly DVOM.
https://www.reflectionsandshadows.com/cabby/static-digi.html

Ignoring your comment about CIS, just read the above and it says don;t read if you have CIS. Do you know why given that my 83 GTi (CIS) has the same hall-effect sender on the dizzy so I would have thought it would work just fine…? I can't see anything in that write up that includes anything Digi specific…

Cheers,

J

The Caveat is:
If your car has CIS, DO NOT READ THIS, it applies to DigiFant and CIS-E (as used on 16V engines).  So if you had CIS-e then it would apply.

I am pretty sure that you could half set it, but the CIS is set at 750 rpm, with it at idle, where as the digi is set at 2250,  removing the CTS sensor after it is up to operating temp, and you have to reset the ECU by reving it above 3K 3 times, then hold it at 2250. so it is a two person operation.