would like a crisper start-up

[Michael]

Engine 2.0 litre Zetec standard except EGR and Pulse air systems removed.
MS1 V3, Wasted Spark
2nd Spark Table is for LPG, but doesn't come into play for this issue.

Car drives very well, I'm very pleased with it, but I have just a small niggle: it starts quite easily, but I would like it better - turn the key, it cranks, I count 1, 2, 3, 4 (slowly) and its away. On the old EEC-IV it would start before I could even think "1" as does my wife's Almera. I keep increasing ASE and Cranking pulsewidth a few points. I've reduced cranking advance from 8deg BTDC to 6deg BTDC. I also reduced the first row of spark bins above 45kpa from 13deg to 8deg BTDC. The ignition changes seem to have helped a bit, but I can't help thinking that the fuelling needs optimising in the start-up area.

The logfiles show that it catches quite easily, then the revs drop back into cranking, this happens two or three time and then its fine.

Would someone mind taking a look at the attached msq and log files and advise me if there is anything obvious I should do?

Many Thanks - Michael.

[elutionsdesign]

Look at the "cranking" indicator in the logs, it is coming on and off as you crank causing the cranking pulse width to go on and off. I suggest that your engine cranks faster than the 300 cranking rpm currently set in Cranking Settings. Try raising this so the cranking pulse width is used more effectively. Also your MAT signal is not looking like its working well, it is very spikey and reading much, much lower than the coolant and I assume this was a totally cold start?

[Michael]

Ok Thanks, I'll up the cranking RPM.
Yes the first one was a totally cold start, for the subsequent logs I'd topped the engine immediately after starting and tried again to et the most out of the cold start opportunity!! Will report back tomorrow morning. - Michael

[john.p.clegg]

My twopenneth
Your cranking rpms vary between 480 and 170 rpm.....how's your battery?

John

[Michael]

Thanks Elutionsdesign and John.

Increased the ranking rpm to 500 which was an improvement, but a bit too far. Brought it back down to 400 and warm starts are a lot better. Now have to wait till tomorrow morning to get a proper cold start.

John, battery is not in its first flush of youth, on the other hand it not ready to be pensioned off yet. Electrolyte sg's are all good, I can leave it a fortnight and it will hold a charge and start up straight away - notwithstanding the present issue.

I'll update tomorrow.

Michael

[Michael]

I've done a few things. First of all following Elutiondesigns suggestions, I upped the cranking RPM to 500 which was a bit much and then lowered it to 400 which improved warm starting, but didn't make much impression on cold starts.

John's suggestion got me thinking, the battery itself is OK, I'm sure. There were issues with wiring which I was never satisfied with since I fitted the Zetec about 5 years ago. First of all and quite relevant the battery feed for the ECU was taken from the wrong side of the ammeter which itself was more of an irritation than a problem as the ammeter always showed a charge. However, it was taken from the battery connection to the starter motor, where it would see maximum voltage drop on cranking. I have now wired the ECU feed correctly from a "clean" battery feed (switched by ignition, of course). The second thing I did was wire in an additional ground from one of the starter fixing bolt to the car bodywork. These two things brought up the voltage shown on Tunerstudio from 11.6 to 12.3 (battery reads 12.4 - 12.5 cold on my DVMs). So now the voltage drops shown on the log files are not so deep, starter seems to spin faster, but cold starts should be better. I think the battery is now eliminated from the equation.

I can see from the log that the engine catches quite easily and the MS hands over from cranking to ASE/WUE, revs drop then goes back to cranking hands off to ASE/WUE and repeats this process three or four times before becoming self-sustaining. I'm sure it should pick up from the intital crank.

My gut feeling, after looking at the logs is to substantially increase ASE, but I'm not confident I'm interpreting the log properly. This is based on WUE being 135% at about 10degC (CLT for cold start was 11degC), but ASE is only about 20%. This suggests to me that I could increase ASE by 15% making it 35% to match the 135% used in WUE. Intuitively this seems wrong.

(Please understand, it does start very reliably, but I am just wanting to get get it that bit sharper - if possible. If this is as good as it gets, I'll stop worrying about it))

Can anyone see anything in the attached msq and log that I should adjust to get a cleaner start?

Many Thanks - Michael

[Yves]

FWIW : 3°C temps outside here, in the garage maybe 10°C (50°F) and ASE is set at 48% in mine. Starts fairly easy and stays running, albeit lopey. I came from around 35-40% where in the last colder outside temps it seemed to wanting to stall

[Michael]

Thanks Yves - makes me want try a higher percentage ASE. Might just do that in the morning and report back - Michael

[nathanhardy]

so what did you find

[Michael]

Sorry about the delay in reporting. I'm still working on it. ASE and WUE enrichment didn't seem to help, so I reversed those changes. On a separate issue I then made some changes to the idle valve frequency (Ford PWM valve), changed the frequency under Idle control to 33 (300Hz), was 70. (See, "Ford Fidle ISCV, 2 wire idle valve frequency measured" under Board assembley) I then found that the Cranking DC had to be changed also had to be changed. Reduced this initially to 20 from 50 which was clearly unworkable. Further reducing Cranking DC to 10 improved startup, presumably by enriching the mixture. Recently as the weather has got colder here (Droitwich, Worcs, UK) I made quite a big increase in priming PW. This also improved cold starting and made hot starts much better. There is still some way to go, further enrichment, I think will improve cold starts, also with extended cranking MS resets as battery voltage dips (New Battery fitted) so I'm going to add some larger capacitors around the 5V reg to hold the 5V line up during transient dips to hopefully avoid the resets. This mod is suggested in the manuals somewhere, although from memory I think it was for V2.2. Should also apply to MS1. Also moving house at present Will post further update ASAP.

Michael

[Brian]

I don't have tunerstudio on this computer to look at the actual tune, but no one has mentioned the priming pulse. If you just want it to catch on quicker and everything else is OK, maybe give it some more priming pulse. I like to be careful with it to not flood the engine with repeated starting attempts, but the extra prime will give you the fuel to wet the valves backside without waiting for 8-10 pulses of cranking pulsewidth to do it.

Edit- You may also want to pull some of the cranking pulse and or ASE to compensate this additional fuel.

[Michael]

Thanks Brian, I agree with your strategy. I'll be in one place for the next few days (between houses at present owing to house move to deepest Wales) so will hopefully be able to get to car and report back before end of week. Thanks - Michael

[mickeymarrows]

Just like Brian I am not in a position to check your tune files on this work PC, but it sounds identical to my problem. It aways started, but always took a few seconds. Sometimes it would start to catch so that I had to keep cranking until it picked up and ran.

My problem turned out to be a corrupted Cranking Pulse table/curve caused by importing the table from an existing tune in degrees F to my tune in degrees C.

What are your Cranking Pulse %'s?

[Michael]

Well, despite making the cardinal error of making two changes at the same time - progress at last!
I was getting occasional MS resets if cranking went on more than 3 or 4 seconds. One could argue that if it started quicker I wouldn't have to crank so long, its a bit chicken and egg I suppose. On this car the battery is in the boot so there's quite a long cable from +ve to the underbonnet fusebox and relays. Although I have fitted a new battery which was an improvement, I suspect there is still a significant voltage drop on the +ve cable. (Now that the MS installation is settled, the car is due for a re-wire and re-plumbing in the spring when I shall address this issue properly). However, I decided to reduce the susceptibility to resets by implementing the mod mentioned above. It is described in the V2.2 Megamanual here:
http://www.megamanual.com/v22manual/mtune.htm
Go to this page and then ctrl +f to find enter reset and it should take you to the appropriate entry.
There is also good stuff on resets here:
http://www.extraefi.co.uk/resets.htm
I re-read the Megamanual notes and checked on the different 5v regulator specs (LM2937ET, LM2940 and 05CC0). My MS1 has the Regulator (U5) which is marked "05CC0" this is abbreviated to fit the part number on the IC, the full name is BA05CC0T and its made, probably, by Rohm. 2937 has very good Low Drop out capability: A regulator, 5Volts in this case, has to have an input voltage greater than the output. A standard regulator like the LM7805 needs 2.5 volts difference making a minimum input of 7.5 volts to produce a regulated 5volts at its output. The difference, 2.5 volts, is used for the internal operation of the regulator. If the input voltage drops below 7.5 volts, then the regulator "drops out" of regulation and the output is shut down. The regulators mentioned above, used in MS, are Low Drop Out (LDO) types requiring less than 1 volt above the output voltage for correct operation. The 2937 performs very well, but can only supply 500mA which is no doubt why it has been replaced in later versions of MS with either LM2940 or 05CC0. 2940 is a very good regulator for automotive applications requiring between 0.5 and 1 volt overhead, but the 05CC0 is just a bit better requiring only 0.3 to 0.7V. So if you are following the V2.2 manual, you only need to change U5, if you presently have an LM2937 fitted, if you have a 05CC0 don't bother changing U5. I did not delete the Zener diode (D12 on MS1, D15 on V2.2) as I doubt that it draws significant current until its breakdown voltage is reached (22v) so I don't think it contributes to the issue of resets caused by a drooping supply. Furthermore, it is there to protect U5 from overvoltage surges so I felt better leaving it in circuit. Similarly D19 (MS1) or D12 (V2.2) is there to protect U5 against reverse current conditions. D10 (MS1) and D13 (V2.2): there is some logic in deleting this as it will contribute a 0.6v drop across its junction. On the other hand it prevents the charge held in the input capacitor (C16 MS1 or C15 V2.2) from leaking back into the 12V rail and depriving U5 of the benefit.
So that leaves the two capacitors either side of U5. The output capacitors (C17 MS1 and C16 V2.2) according to the 2940 Spec, as long as they are greater than 22uF can be of unlimited size; the 05CC0 spec suggests an upper limit of 1000uF. The specs do not limit the size of the input capacitor (C16, C15), but experience suggests that this can be quite big.
With all this in mind I was then quite happy to replace both capacitors with 470uF 25V Electrolytics that I had on hand.
The second thing I did was to increase the Priming PW by about 0.5ms across the board.
Starting is now greatly improved, the car has stood outside for 6 days in frequently below freezing conditions after these changes it started very quickly with no resets. There's a little way to go yet, after monitoring this for a few days, I'll see if it needs another 0.5ms of priming PW.

For MickeyMarrows my Priming PW table:

Priming PW.jpg

Also attached is latest msq and a datalogs of the cold start.
Any comments or advice welcome. (hope this did not ramble on too much!)
Thanks Michael.

[mickeymarrows]

Its been a while since I played with MS1 (I highly recommend the upgrade to MS2, especially as MS1 will be obsolete soon) but it appears to me that you have zero cranking advance.. the log shows it and its zero in the tune file too...

[Michael]

I set the cranking advance to zero some time ago, when it was sluggish on the old battery and I got resets which I mis-diagnosed as too much advance kicking back against the starter. I've left it like that as it seems to work well, but I may well re-visit it in the near future. I need to think about it as the propagation of the flame front is at a constant speed almost regardless of engine speed ( I know this is not perfectly true, but its a fair approximation for sub 1000RPM idle and cranking) so you have to consider time as well as deg BTDC. Take an idle speed of 800RPM with a typical advance of 10deg BTDC, if you do the maths that works out as: 13 revs per second; 0.075sec per revolution; 1degree every 0.0002 second. So 10 deg BTDC allows 0.002, or 2 milliseconds, after the mixture ignites to develop almost maximum pressure. (Maximum is just after TDC). An assumption here is that 10 deg BTDC is a fair setting at 800rpm, the engine idles well at this setting and it approximates to the manufacturer's spec, therefore 2 milliseconds is a fair propagation time for this engine to develop meaningful cylinder pressure. If you now move to cranking speed, say, 250 RPM. This is 4.16 Revs per second; 0.24 sec per revolution; 1 degree every 0.0007 second. If you have cranking advance set at 10deg BTDC, which would not be unusual, there would be 0.007 or 7 milliseconds from point of ignition to TDC. That's over 3 times as long as at idle so timing kickbacks are to be expected as maximum cylinder pressure would be achieved before TDC. Given that the idle propagation time is reasonable then a suitable Cranking advance would be a tad less than 3 Deg BTDC as this will allow 0.0021, just over 2 milliseconds propagation time.
This is just thinking out loud, I am more than willing to bow to superior wisdom/experience in these matters so please offer corrections or advice.
Now I've got that off my chest I'll try it next time I'm out in the cold and let you know.
Best Wishes - Michael

[mickeymarrows]

Me thinks you may be over thinking this Michael, however your thinking makes perfect sense.

However, if engines always fully complied with the laws of physics we wouldn't need to tune, we could simply write algorithms to populate the variables for us

It may well be that zero cranking advance is correct for your engine, but looking at a log with only the RPM and Spark Angle traces visible it doesn't look right (to me at least).. you end up with a scenario where at 395rpm spark advance is 0 yet at 405rpm its 8 degrees, it drops below 400rpm and spark advance drops back to zero (I don't have your log open so forgive me if the figures are not spot on, purely for illustration)

From experience I know that my 16v starts perfectly on 8 degrees, but my V6 prefers 10, at 12 degrees the V6 starts violently which suggests to me that it is just at the point of kickback.. At 4 degrees the V6 takes a second or two to fire..

[Michael]

All points well taken! I do believe if it works don't fix it! I monitor the startup performance over the next few days and might put in a few degrees of advance as I set it to zero on a false premise. I see your point about timing swinging from 0deg at cranking speed, then shooting up to 8 or 10 deg as soon as the cranking threshold is exceeded.

Michael.

[john.p.clegg]

Keep it up boys,I'm sure there are plenty of us lurking waiting for that magical msq for four cylinder fords...

P.S. are there any Ms1 maps for Zetecs etc stored anywhere?

Thanks
John

[quan3165]

My last big block Chevy comes to mind here. I couldn't get the thing to start crisp without 15 degrees of timing during cranking. I also am reminded of the NA race car guys who swear locking out your timing at 30 degrees is the best way to prevent erratic timing from mechanical and vacuum advance combating each other all the time. Their stuff starts right up with enough cranking amps. 12+ degrees during cranking might be worth a shot if you can get it to turn over at that.