DG Hot Start Problems

[zoona]

I am going to show my vw/distributer/carb lack of knowledge again... I didn't realise that's how the rev limit was set.

Anyway, the one I bought was non limited it seems: https://www.brickwerks.co.uk/rotor-arm- ... w-use.html

Hoping to have some time to check the push rod this afternoon. Then rotor arm safety that. I want to do it one thing at a time rather than chuck everything at it and the problem guess away.
It's just a pain to reproduce, having to get it hot etc


I am rapidly losing faith in it, and I want to trust it again...

(Thanks for the offer of a fuel pump)

What pressure pump would I need to buy if I go down that route? I see there are 1.5-4, 4-7 and 7-10... Lower I am assuming... And then do I need to regulate it?

[zoona]

I have just taken out the fuel pump push rod, and found the following things:

• It is 99.5mm when i think it should be 100mm
  
• The fuel pump spacer was full of a mixture of oil and fuel. It was still quite viscose, like it was mainly oil, but it definitely smelled of fuel as well.
  

I have cleaned it up, and cut two new gaskets as the old ones looked like they were older than the fuel pump (which i believe was fitted not long before we bought it).

Questions:

• Is a 0.5mm drop in throw on the fuel pump lever enough to make it weak? I'm not sure what the throw should be, but it could be a significant difference i guess
  
• Is the spacer being full of oil/fuel an issue? i think there is an easy path for oil up on the rod, but i don't think there should be a smell of fuel there?
  

[DoubleOSeven]

The fuel pump push rod being 0,5mm short. I’m not convinced this is your issue, it would still pump. The gunk in the bakerlite base, no big deal, most are like this. The gaskets just stop the oil coming up from the engine, not related to fuel supply.
I had another thought. When it’s hot take the pancake air filter inlet off, is the choke flap in the carb still closed? It should be open. This would prevent hot starts. This is what the repair call our guy was effectively
working on, albeit putting the carb cleaner in there rather than giving it more air. I had this issue on my Mk2 Golf Pierburg.

[zoona]

Ok, I'll try that next time I am looking at it

Off on a non camping holiday today, so will be a while before I can try it...

[liamd]

Ok- so I've been having what looks like the same problem with my 1.9DG.

It has been the bain of my life for a while so I want to set out the diagnostic work in case it helps anyone solve the riddle- it has been inconclusive to date but I will be going to a specialists pretty soon and will update the forum as to what comes out of it.

However- in the meantime...

The problem started when coming off drives where the engine gets fully up to temp, you attempt to start it between 20mins to 1h30ish after switching off. Importantly it would always start initially- but would die after 20-30 seconds. 

Once it has done this it will not start using any normal starting method- having throttle open while cranking or relying on the idle or somewhere in between or cold start process.

Each and every time it would start initially after being switched off and then die- importantly it will not start again without intervention until cool.

The choke is totally off once warmed up in this situation as it should be.

The two methods that have worked to get it started in order of effectiveness are:

- holding the choke closed so as to leverage the vacuum of the engine cranking to draw fuel - releasing once catching.
- chucking some petrol down the primary venturi - although rarely in the first 20 mins.


So it's the fuel pump right?

I'm not convinced. Although it may be a contributing factor.

You see, at the point at which this happens I may have been driving for 5 hours at 70mph without a hitch- clearly the fuel pump can supply peak demand when the engine is hot.

So it's vapour lock? 

it's certainly possible BUT the engine temp gauge clearly shows that the operating temp is in the normal range- fuel lines aren't resting on the manifold or anything silly. See note below about modern fuel.

Perhaps the float pin is sticking sending the fuel back to the tank?

This has been my central assumption.. essentially that the warm engine ticking over doesn't use much fuel and that allows the float bowl to fill (perhaps this only happens occasionally if the pump is weak). maybe some but not all the petrol in the float bowl evaporates after turning off and for whatever reason the float needle stays lodged sending fuel back to the tank upon re-start. Once you burn off the little bit remaining you essentially have no fuel supply.

The obvious the way for me to test this is to disconnect the return pipe when this happens and see if fuel is being returned to the tank- however the screw for the pipe is rounded off so hasn't been possible at side of the road. Additionally the float bowl may well actually be full- it isn't possible to tell as far as i'm aware.

To add to the mystery on two occasions it has died while idling / in traffic and been similarly difficult to start (one was on the downhill stretch that passes Stonehenge in heavy slow traffic.. yes- convenient!!) as i actually had a good long downhill I tried to bump start it down about 50% of the hill and absolutely nothing. Had to chuck petrol down the venturi- and that didn't work for a good hour or so. that was before I learned the forced choke method.

The second time I was outside my house trying to simulate a hot start- I let the van idle for 20 minutes, and slowly but surely it began to cut out. I took the pan off the carb and asked the mrs to pump the accelerator... fuel came out into the primary. not just one pump but enough for a few pumps of fuel. There is clearly still fuel in the carb at the point at which the idle is cutting out at least. This doesn't quite fit with the float needle hypothesis- unless I'm missing something. 

Despite what the temp gauge is saying- to my senses the engine bay seems hotter than it should be- even though the radiator appears to be working and isn't too old. It has the right coolant in it- but possibly too higher percentage leading to poor cooling.

The plugs are inconsistent on inspection, Importantly they don't look like the last set of plugs- which looked spot on.

In short something has changed and I wouldn't be surprised if the engine were getting a little hot.

The car has fresh oil, filters etc.

I have not split the carb- but I have cleaned with carb cleaner. There could be crap in the float bowl.. but the failure conditions seem too consistent for that to be the case and even with the vehicle stationary (i.e not shaking around) I get the same issue.

Other than this being a case of covert overheating with it's many causes- this leaves a hard to spot vacuum leak that could be having secondary effects on timing etc.


So my questions...

Of the hypotheses above i reckon the float needle getting stuck is the strongest- but I can't understand why once the engine is cold it will start again. is it possible that when the full vacuum force of the engine is applied to the carb when cold and choked it is sufficient to suck the stuck float pin out? Or could there be contraction of metal components contribute to it being dislodged?

I have also assumed that the reformulation of petrol (ethanol) since the t25 was in production means vapour issues are more common even within what would historically be normal operating temps. Am I right in thinking that the temp sender is actually taking it's reading directly from the engine block / oil and not the coolant itself?

Any other thoughts / condolences welcome!

[Mocki]

If you suspect the fuel is "running back" to the tank , the fuel pump is had it . The pump is a diaphragm pump, so basically it’s a rubber disc that flexes as the push rod moves up and down , causing vacuum and sucking , on the down stroke as it were , the disc seals the route back to the tank to maintain the vacuum. 

so if it’s running back , the diaphragm is failing , which would also tally with the excess spinning over to start when hot . The temperature of the rubber being a key factor , especially if it’s already porous or cracked 

also check the cone filter in the carb inlet spigot , it’s often overlooked 

[liamd]

Hi Steve,

Thanks for your reply- which I believe is spot on from my further investigation.

I will set out the fail conditions today and the tests that I did- all of which point to your analysis being correct. New quality pump going on tomorrow.

So today I drove up to the VW engine company in Keston and had a chat with Adrian. This is about a 30 minute drive for me through london traffic.

I left the engine running the whole time whole he looked over the engine to check for anything noises pointing to a bottom side issue. He believed that mechanically it was running relatively smoothly. By the way- they have an amazing new place, and are very friendly.

The fan kicked in a few times but the van kept ticking along fine- as it did on the 30 minute drive home.

I parked the van on the flat when I returned home- left the engine running for about 5 minutes. Again the fan came on, but this time slowly but surely the van started to die.. I blipped the accelerator a little to see if it was idle specific- but it just eventually cut out.

I believe the critical difference between the idling at the garage and at home was that the cover was off the engine which will have very slightly reduced the temperature in the engine bay while at the garage.

While the engine was still hot I took the fuel supply pipe off the carb and disconnected the generator. I placed the fuel pipe into a plastic bottle and got the mrs to crank the engine. Nothing. The fuel pump was not delivering any fuel.

I then checked the transparent fuel filter by the fuel tank and can see that there is fuel in it.

Finally I reconnected the hose to the carb and leads to generator and cranked it with the choke closed. The engine caught briefly (at which point I took the choke off a little) but did not start drawing fuel so as to continue running.

Another observation that would support the fuel pump diagnosis is that this has been getting worse over time AND first started to reveal itself in traffic going downhill. It this state the engine isn't working hard, but the fuel pump is fighting gravity- exposing any little breaks in the diaphragm I imagine.

Finally it might be worth mentioning a couple of other things- a) the current fuel pump is the tinny low quality type and likely easily gets too hot internally b) There is a possibility that through topping up over the last couple of years the proportion of 'coolant' to 'water' is no longer within tolerances- it’s probably mainly g13.

The reason I mention this is because, as I understand it, concentrated coolant doesn't have the same thermal capacity as when properly mixed with water. This means it neither picks up as much heat from the engine on it's passage through it, nor emits as much heat when it goes through the radiator. As the engine temp gauge takes the temperature of the coolant NOT the engine block there hypothetically a state where the engine is running hotter than any of the sensors show.

What I'm saying is- certainly there is an issue with the fuel pump- but that excess heat may have revealed this sooner than otherwise and that this excess heat could be hard to identify using gauges alone.

Just a thought.

Thanks again for your reply steve.





 

[zoona]

Update on my problems...

I have replaced the fuel pump push rod. Made no difference.

It has been into the garage to get the carb set up properly. I tried, and my patience failed.

While there, the tip came off the distributer rotor arm, so that got replaced.
It was also due an emissions test, which I discovered needs one every year.
840chf later, it was back in the road and running well - all except for the hot start issues... I need to stop looking at the prices.

I did speak to the garage who did this work (they have a dokka as their garage runabout, so happy they know what they are doing) and was told that I can fit an electric pump.

So there is one on the way to my house as we speak. Went for a pierburg, with relay/wiring to make it not burn the car in a crash.



This weekend, it managed to get over the Grimsel Pass which i was very happy about (to see this)



but then today it conked out in Täsch, on the way to Zermatt, in a particularly embarrassing and inconvenient place.
I will be very pleased to get this new pump on and see how it goes.

[zoona]

Just broken down after stopping for fuel in Obergoms.
Sat outside a Garage waiting for it to cool down.

It started up after a few seconds. Then cut out 100m down the road.

I am going to fit the electric pump as soon as I get home.
If that doesn't solve my problems, it's going in the bin.

[maxstu]

Heart goes out to you Ian. It really dampens the camper van spirit when you cant get to the bottom of a very niggling issue.
I had similar problems in my Auto DJ. Constant stalling at most inappropriate places. Worst one entering a roundabout towing a caravan. Finding N while freewheeling in a panic before trying to restart got me sweating.Finally solved by a very experienced old skool VW trained engineer. Tested my patience to the limit though.
I cant recall all your postings about this problem. But what hot start procedure are you using? Throttle halfway l was told. And always worked for me.
Are you certain it is fuel supply related? I would re-read the information given about the choke flap above. Or perhaps carb float height?
Also if diaphram failure in fuel pump. Any new fuel pump would instantly solve your problem.
You did test coil too?

[zoona]

we are back now, and it started when it had cooled down, but still required some start spray to get it going.

Until now, i was convinced that it was the pump, but having taken the top off it while trying to start it, there was definitely fuel in it. Which makes me worry that this isn't the problem.

Nothing makes it work with a hot start. tried everything - apart from cooling down a bit and start spray

So, as you suggest... I think I need to look at other things as well.
(The carb was checked by the garage I took it to, so i think that is ok... but if/when i bring it back to the UK to get the roof sorted, I will get somebody to go through everything - i just don't have the time at the moment.)

[zoona]

I have fitted an electric pump. And also replaced the bit of hose that I couldn't get to previously (the hard line into the engine bay)

So the installation all went well - i.e. I had all the bits and didn't have to order the things i forgot, which is what usually happens...

Now when I start it, it seems to take quite a long time to start - i.e. 2-3 seconds of cranking before it fires up. That felt like it was too long, and I was expecting it to be pretty much immediate.
I think i just need to drive it and see if this is the "new normal" for want of a better phrase.


I followed the instructions with the pump, and have this set up now.


                                                      Heater
                                                       Pump
                                                         |
Carb-------------NonReturnValve--filter--pump--filter---------Tank
  |                                                             ^
  ---------------------------------------------------------------

It said that the filter between the tank and pump should be a course one, and then a normal one after it. I didn't have a coarse one, so just used a normal one in both places. I'll get a coarse one now though, as i assume it will put a bit more strain on the pump?

It is mounted directly on the chassis (as per instructions) and it is pretty noisy inside, so I'll get some rubber isolating mounts as well.

I just need to drive up a mountain and see if it conks out when I get to the top...
   

[Mocki]

Which pump did you get ?
i have a cheap farcet pump , and it is before the inline petrol filter , as it has its own filter in its inlet .
there should always be enough fuel on the carb to start the engine before the pump catches up anyway .

[zoona]

Which pump did you get ?

A pierburg. I think it is a bit much for this application, but a german "vw specialist" assured me it was right.
I think I should have ordered a facet and waited a bit longer for it to arrive. I may still swap if it causes problems.
Like this one: https://www.auto-doc.ch/pierburg/965133 ... 98QAvD_BwE

there should always be enough fuel on the carb to start the engine before the pump catches up anyway .

That's what I thought... I think I just have to use it a bit and see what it is doing.

[zoona]

I've been running it for the last couple of days, and it has started perfectly every time. Not sure what was going on there.

Need to do some stress testing next...

Thanks for everybody's help on this one. Will report back after our next long trip. Hopefully not from the side of the road.