Cold start - how long does your engine take to catch?
Discussion
Hi all,
After thinking I had the cold start sorted on the new Emerald system, this week has proved me a bit wrong now its in -ive numbers. When very, very cold this week when starting up in the morning, the engine seems to take around 10 cranks (or more) to even think about firing up. Its almost as if its got bugger all fuel, even after waiting for the fuel pump to prime fully. When its been running with a touch of throttle for 5 seconds, it idles absolutely spot on and drives without any problem whatsoever and when restarted when anything but very cold, its back to a very quick start and works perfectly.
I was just wondering how long it should take when very cold and what I might be able to do to improve the start? I've bumped up the cold start injector priming to 14ms to get a good squirt of fuel in there before it first sparks. I've also set the very cold idle to 1250rpm with a bit more fuel and air from the stepper. This seems to make it idle just fine, but its the very cold start that sucks.
In case it makes any difference it started doing it this week when its been very cold, but I also had the fuel filter changed last week, is there a chance that the fuel pressure is taking longer than normal to build up when its been left over night perhaps? - I'm only taking a stab in the dark as I'm not familiar with the fuel system but its the only thing thats changed apart from the weather...
Any help would be greatly appreciated!
Nick
After thinking I had the cold start sorted on the new Emerald system, this week has proved me a bit wrong now its in -ive numbers. When very, very cold this week when starting up in the morning, the engine seems to take around 10 cranks (or more) to even think about firing up. Its almost as if its got bugger all fuel, even after waiting for the fuel pump to prime fully. When its been running with a touch of throttle for 5 seconds, it idles absolutely spot on and drives without any problem whatsoever and when restarted when anything but very cold, its back to a very quick start and works perfectly.
I was just wondering how long it should take when very cold and what I might be able to do to improve the start? I've bumped up the cold start injector priming to 14ms to get a good squirt of fuel in there before it first sparks. I've also set the very cold idle to 1250rpm with a bit more fuel and air from the stepper. This seems to make it idle just fine, but its the very cold start that sucks.
In case it makes any difference it started doing it this week when its been very cold, but I also had the fuel filter changed last week, is there a chance that the fuel pressure is taking longer than normal to build up when its been left over night perhaps? - I'm only taking a stab in the dark as I'm not familiar with the fuel system but its the only thing thats changed apart from the weather...
Any help would be greatly appreciated!
Nick
Do you have a "first start of the day" feature? On MegaSquirt with this enabled, pumping the throttle twice before cranking roughly doubles the amount of fuel (i.e. twice as much as normal cold starting, which is already pretty generous). Worth seeing if you have something similar; if not, try increasing the cold start enrichment at the bottom of the temp scale (don't know Emerald but I assume it is mapped on coolant temp).
Hi Pete,
That's correct, it is mapped on coolant temp but then has a fine tune based on air temp as well. For the cold start, there isn't a 'first start of the day', the nearest thing to it is the initial injector prime which gives a single squirt of fuel as soon as engine rotation is detected. This is currently set to 14ms (from 12 when set up by Dave @ Emerald). There is then the cranking enrichment % table which can be used to alter the fuel for the first 1, 2, 3, 50 and 100 engine turns (adjustable) at varying coolant temps in order to decay the startup enrichment over time. The manual says this is intended to combat fuel dropout at low induction speed on startup so I'm not sure how much of an effect this would have. Other than that, I'm not sure what to try... Perhaps just keep bumping up the inital injector prime and enrichment % until it has an effect? - though I don't want to end up flooding the thing!
That's correct, it is mapped on coolant temp but then has a fine tune based on air temp as well. For the cold start, there isn't a 'first start of the day', the nearest thing to it is the initial injector prime which gives a single squirt of fuel as soon as engine rotation is detected. This is currently set to 14ms (from 12 when set up by Dave @ Emerald). There is then the cranking enrichment % table which can be used to alter the fuel for the first 1, 2, 3, 50 and 100 engine turns (adjustable) at varying coolant temps in order to decay the startup enrichment over time. The manual says this is intended to combat fuel dropout at low induction speed on startup so I'm not sure how much of an effect this would have. Other than that, I'm not sure what to try... Perhaps just keep bumping up the inital injector prime and enrichment % until it has an effect? - though I don't want to end up flooding the thing!
Hmmm... I'd be inclined to look at the ign advance side of things (but watch out for big backfires 
)
One thing I do remember from discussing the spec of my tornado chip with MA is that he builds in a start 'delay' to enable oil pressure to build over a second or two cranking before fuel is delivered. Might not be such a bad thing if it doesn't catch immediately?
If it helps, mine (still on Lucas/tornado
), is showing in the range 11:1/12:1 AFR immediately on start up (catches quickly but won't accept any load), then quickly moves to ish 14.6:1/15:1 where it will drive
)One thing I do remember from discussing the spec of my tornado chip with MA is that he builds in a start 'delay' to enable oil pressure to build over a second or two cranking before fuel is delivered. Might not be such a bad thing if it doesn't catch immediately?
If it helps, mine (still on Lucas/tornado
), is showing in the range 11:1/12:1 AFR immediately on start up (catches quickly but won't accept any load), then quickly moves to ish 14.6:1/15:1 where it will driveI do agree that a longer crank might have its benefits, but my friend came out with me today and commented that it 'almost didn't start!' this morning again. This was after I let the fuel pump prime, switched ignition off, switched back on and let it prime again and cranked with my foot on the throttle a bit. Again, it cranked for a while then spluttered and burst into life and idled just fine after a couple throttle induced 2k rev seconds. I'm just curious as it wasn't actually that cold this morning (it was around 11am and this is in an underground carpark so its a couple of degrees above outside temp). It didn't do it at all before I got that fuel filter changed and since it's done it every morning! 
thenick said:
I do agree that a longer crank might have its benefits, but my friend came out with me today and commented that it 'almost didn't start!' this morning again. This was after I let the fuel pump prime, switched ignition off, switched back on and let it prime again and cranked with my foot on the throttle a bit. Again, it cranked for a while then spluttered and burst into life and idled just fine after a couple throttle induced 2k rev seconds. I'm just curious as it wasn't actually that cold this morning (it was around 11am and this is in an underground carpark so its a couple of degrees above outside temp). It didn't do it at all before I got that fuel filter changed and since it's done it every morning!
What condition is the battery in Nick? If its a bit marginal and cranking amps are a bit below par then it could be that the coil drivers are not getting the feed they need as all the current is being drawn by the starter. Cold weather will depress its output further; maybe try and see if it's better off a jump? Not at all related to the fuel filter but p'haps a coincidence...chrispitman said:
Hi Nick
Are you still using the original coolant sensor, as in the emerald manual it says you have to calibrate it, the software has default or user defined where you can calibrate the sensor and it maps a curve, if this was out could this be causing some of your problems.
I'm using a brand new coolant temp sensor from Emerald but its exactly the same as the old one really. This type of sensor apparently doesn't need calibrating as the Emerald is already programmed to use it. I think you only have to calibrate if you're using a different kind of sensor. It seems to be reading pretty sensibly and follows the TVR gauge (except the gauge reads about 10 degrees lower), though I know that the gauges are often out by this much so I trust the Emerald sensor!Are you still using the original coolant sensor, as in the emerald manual it says you have to calibrate it, the software has default or user defined where you can calibrate the sensor and it maps a curve, if this was out could this be causing some of your problems.
Edited by thenick on Thursday 21st February 23:02
Pupp said:
thenick said:
I do agree that a longer crank might have its benefits, but my friend came out with me today and commented that it 'almost didn't start!' this morning again. This was after I let the fuel pump prime, switched ignition off, switched back on and let it prime again and cranked with my foot on the throttle a bit. Again, it cranked for a while then spluttered and burst into life and idled just fine after a couple throttle induced 2k rev seconds. I'm just curious as it wasn't actually that cold this morning (it was around 11am and this is in an underground carpark so its a couple of degrees above outside temp). It didn't do it at all before I got that fuel filter changed and since it's done it every morning!
What condition is the battery in Nick? If its a bit marginal and cranking amps are a bit below par then it could be that the coil drivers are not getting the feed they need as all the current is being drawn by the starter. Cold weather will depress its output further; maybe try and see if it's better off a jump? Not at all related to the fuel filter but p'haps a coincidence...
I'll have a better look at the weekend as I want to put in some B7 spark plugs and tidy up the engine bay a bit so I'll check for a spark and see if the injectors are firing straight away...
How's your one going by the way Gary? Got it fired on the Emerald at all yet?
thenick said:
Pupp said:
thenick said:
I do agree that a longer crank might have its benefits, but my friend came out with me today and commented that it 'almost didn't start!' this morning again. This was after I let the fuel pump prime, switched ignition off, switched back on and let it prime again and cranked with my foot on the throttle a bit. Again, it cranked for a while then spluttered and burst into life and idled just fine after a couple throttle induced 2k rev seconds. I'm just curious as it wasn't actually that cold this morning (it was around 11am and this is in an underground carpark so its a couple of degrees above outside temp). It didn't do it at all before I got that fuel filter changed and since it's done it every morning!
What condition is the battery in Nick? If its a bit marginal and cranking amps are a bit below par then it could be that the coil drivers are not getting the feed they need as all the current is being drawn by the starter. Cold weather will depress its output further; maybe try and see if it's better off a jump? Not at all related to the fuel filter but p'haps a coincidence...I'll have a better look at the weekend as I want to put in some B7 spark plugs and tidy up the engine bay a bit so I'll check for a spark and see if the injectors are firing straight away...
How's your one going by the way Gary? Got it fired on the Emerald at all yet?
How complicated is the cranking setup on that ECU? The Megasquirt one is slightly wierd and has several modes available for cranking, some of which are completely different to normal running. It would be the easiest thing in the world to get the dwell angle wrong, or get twice as much fuel as expected (or half as much), or get the ignition timing a long way out. Since this only happens during cranking it might only show up as a reluctance to fire.
As far as I'm aware, there aren't a selection of cranking modes as with megasquirt. You just set it up with its various parameters and it does it's thing...! The map hasn't been changed since it started doing it though, except the tweeks I've tried, but I've tried all three maps I've got on there (largely the same though at low revs/crank) and they produce the same result...
The fuel pressure idea could be something to look at? It does seem like it has no fuel pressure when it first tries to crank despite the fuel pump doing its thing as normal. How would I check the fuel pressure reg if this could be causing problems? Also is there any other way the pressure could be leaking out without me smelling any petrol?
The fuel pressure idea could be something to look at? It does seem like it has no fuel pressure when it first tries to crank despite the fuel pump doing its thing as normal. How would I check the fuel pressure reg if this could be causing problems? Also is there any other way the pressure could be leaking out without me smelling any petrol?
Nick
You say you have a new battery, have you measured the voltage on it when you are cranking the engine over, as a cold battery and engine will require more current from the battery when starting and if it can't supply the voltage will drop. There is a Battery Voltage Compensation setting which efects the injector timing by adding longer duration vs lower voltage.
You say you have a new battery, have you measured the voltage on it when you are cranking the engine over, as a cold battery and engine will require more current from the battery when starting and if it can't supply the voltage will drop. There is a Battery Voltage Compensation setting which efects the injector timing by adding longer duration vs lower voltage.
Gassing Station | Chimaera | Top of Page | What's New | My Stuff