Injector leak back results(hot start problem)

Hi,
Sorry if this is a long post, but this issue has been nagging me for quite a while now.

Car: 06 Accord icdti with 75K on it


background;
This issue started out as an intermittant starting problem and finally became a 'classic' hot start issue several weeks ago.
When the engine had been given a good run to get fully warmed up, if it was then switched off & left for 10 minutes or so it became reluctant to start. Sometimes it had to be cranked for 10 seconds or more before it fired (worst was about 25secs of cranking), until then it showed no indication of wanting to fire.

Current problem;
lazy hot start. The engine always fires within 1-2 secs from cold. When hot it takes a few seconds to fire & when it does it shudders (or splutters) as if it has only just 'caught'
When it runs it is fine, no loss of power, no smoke, returns very good mileage(about as good as I have always seen).

tests & work done;
No DTC's are reported. When I pull a sensor connector to check then the relevant DTC is reported.
Its had a crank sensor replaced as this did appear a couple of times on the DTC's in the past.
No smoke is seen when the engine does fire, & the tacho shows movement while cranking the engine
IMRC EGR MAP sensor etc all checked & where found surprisingly clean & no signs of anything sticking, all sensor conectors checked & found to be good.
Glow plugs removed & checked - Perfect & equal to each other, impedance & current draw for each was almost identical, they all heated showed the same heat patterns when tested out of the engine.
The battery was replaced a few weeks ago as it was showing signs of age(believed to be the original). This made no difference whatsoever to the starting issue.
It's had a new genuine fuel filter fitted in the last couple of weeks. This did improve the issue quite a bit but not completely cure it.

I did borrow a snap-on diag unit which had a test pattern for the fuel system. This consisted of leaving the engine at idle & it then set fuel pressure targets e.g. 20-40 bar, 40-60 bar etc. It hit the middle value all the way to 1200-1400 (gave result of 1300) but then failed on the final one when it was expected to give 1500bar where the system could 'only' manage 1300 bar, the same as the previous step. I'm not entirely convinced this is a fault as it seems to expect a lot from the system if it is only idling during this test?
When driving it hard to load it it feels like there is no loss of power (I've owned the car for several years) which suggests the fuel pressures are good enough at higher revs/loadings.
This all leads me to think it 'may' be a fuel pressure problem when cranking only?

I did a leak back test (several) & got the followng:
runing at idle for 4 minutes(results are proportional when I ran it for 2min)


Trying the idle 2mins, then 3 lots 30 secs at ~3k with 30 secs idle between each


While Inj 1 is not particularly good, it seems it falls within the oft quoted rule of no more than twice the lowest?
I don't know the acceptable values to decide whether Inj1 is the cause, or simply part of the problem :unsure:

questions;
Does anyone here know the actual leak back amounts that are considered within HUK specs?
cranking speed acceptable range?
Actual measured fuel rail pressure (using a gauge on the rail, not ECU reported) during cranking?

Anyone got any further suggestions to try?

Your injectors aren't the problem mate, I would bet money it's a slow operating EGR valve. It's a classic problem on the facelift with the vac operated valve. You can't clean it, try sourcing a replacement from a Civic or facelift Accord.

F6HAD said:

Your injectors aren't the problem mate, I would bet money it's a slow operating EGR valve. It's a classic problem on the facelift with the vac operated valve. You can't clean it, try sourcing a replacement from a Civic or facelift Accord.

Click to expand...

Thanks for the reply fahad.
Dont know why I typed EGR as it was one thing I didnt remove :blush:
Guess what? since posting the car has been 200 miles, stopped & restarted about 10 times & always started perfectly...but in the past this fault has cleared only to reappear again.

When you say slow operating I presume it doesnt reseat itself fully (or quickly enough)?
Whats the impact from (temporarily) clamping the vac pipe to see if the fault re-occurs while its clamped?
My thinking being the vacum is used to pulll the valve open & an internal spring is used to return it to closed, then if it fully reseats overnight & the EGR is disabled then the fault should never occur?

Sorry for more questions, but you got me thinking now & I'd like to try & verify the problem before throwing part(s) at it

It's actually a diaphragm rather than spring loaded like the pre facelift electronic version.

I suspect it's not fully closing as the diaphragm internals (this is my guess based on a few examples I've seen now) expand with the heat. Eventually the valve fully closes again, allowing you to restart the car.

See how long you last with your current fix. If it does it again, I would suggest EGR valve.

F6HAD said:

See how long you last with your current fix. If it does it again, I would suggest EGR valve.

Click to expand...

Thanks again for the reply.
I'll try my bodge then to see if it cures it when the fault returns, or maybe just use a mityvac to pump it shut when I get another difficult start & see if it cures it.
I'll post back when the fault is deemed to be cured...

Let me start by apologising for the massive delay in posting the outcome to this.
I've since sold the car (to a friend) & changed to a Jag XF & only remembered this the other day.

I tried the trick of clamping (disconnecting) the vac pipe from the actuator & this seemed to stop the fault but I never had the opportunity to let the fault develop then force the actuator closed with a mityvac.

Anyhow, the opportunity to vist a friends garage came up so, after cleaning the EGR (see DIY thread) the fault has never been seen since so I think its safe to say it was the problem.

When the EGR was removed & operated with a mityvac, it was obvious that there was a slight squeak just before the valve reseated. It was blasted clean using a shot blaster but loaded with coconut shell(?!) so it wasn't too severe

many thanks to F6HAD for pointing me in the right direction.

Spud

Thanks for updating us mate, and no problem - I do get it right sometimes

I do like the XF, nice cars!!

F6HAD said:

Thanks for updating us mate, and no problem - I do get it right sometimes

I do like the XF, nice cars!!

Click to expand...

You were spot on
I did look seriously at the 8th gen, but I needed an auto this time & the 8th gen auto didn't give me that big grin feeling. As I had to clear out some stuff from the garage to fit it in I decided the Jag would also (just!) fit. When I compared the 2 I preferred the XF.
Now if I could have got an auto version of my old 7th gen...

Well an XF is a big step up from an accord in pretty much every way so not really a fair comparison..

An idtec auto with a nice DPF off stage 2 would have been just the ticket