That First Start of The Day

[stuck]

Ta

[Manxey]

Well I've had a play, albeit only one day... but results are significant, so I thought worth posting.

Firstly I sat and thought for a long while, V8's tend to warm up pretty quickly, so rather than wast the first 17seconds with a decaying fixed map, and based upon what Dave above says about being able to run properly immediately on WUE, reduced this to 5 seconds. I am thinking of putting this down to 1 second at a later date and see what happens

To combat the long-ish starts in the morning, I changed the 'Cranking / Prime table' settings a bit, firstly I changed the pump prime to 'Always', changed the 'Fire priming pulse' from 'power up' to 'Twice', and adjusted the Cranking/Priming PW table' figures to give a bit more fuel whilst trying to start. It's really the first tim eI have looked in such detail and things have 'clicked' about Crank/ASE and WUE. Realising the Cranking/Priming PW table' is fixed and the Fuel VE table is ignored completely in cranking mode makes complete sense, but it also means that the figures have to be best guessed over time, and too little fuel = long crank, too much and a cloud of black smoke after starting.

Anyways, a quick SS of the resulting Cranking / Prime table:

The cranking figures I had were about 60% of what is in there now, I used Neil's figures from above, as it made sense that mine needed to be richer due to longer cranking times.

So first one down, moving on to the 'More cranking settings', not much to see here, I set the flood clear figure a bit higher, and left first start enrichment on and at 30%, I may try turning this off in future, but for the moment it is on and OK.

Next was 'After Start Enrichment', the only thing I did here was reduce the 'ASE TOTAL TIME' to 5 seconds, as the engine should be able to run purely on the WUE table, but 5 seconds, decaying, seems a fair figure. All the other settings were left as is, I think the ASE table is pretty standard, but as said, only active for 5 seconds

Next was the WUE, logic dictated that 17 seconds of ASE was probably masking the fact my WUE was too weak, so as a starter for 10 I picked up Nige's WUE figures from his most recent TunerStudio graph : http://forums.lr4x4....10

Another reason for believing that WUE was too little, was after ASE had stopped, pressing the throttle pedal resulted in a massive bog-down and then the engine would rev up -weak mixture symptoms in my book (and hopefully everyone elses!)

So... here's what I ended up with, including the figures on the RH side:

It looks complicated, but basically you are only interested in a few points on the graph, grab a point and drag it up or down and listen to the engine, you will eventually find a happy medium where it sounds best, mine ended up pretty similar to Nige's in the end in all honesty, but it is definitely much better for it, it actually drives nicely when cold(!)

Oh and that starting? It turned over 3 times, fired, idled at 1100 for a sec, then dropped down to 950 as the PWM valve closed a little. When ASE finished, it remained running with no engine note change, probably due to the nice way MS decays the ASE over time rather than switching it off immedaitely after a few seconds. During warm up and WUE tweaking, when pressing the throttle there was no hesitation at all.

Overall I'm a pretty happy bunny, hope these figures help other people too, I'll probably tweak them again, but I think I must be 90% of the way there now.

BTW, mines a 4.0 P38 engine, Kent H200 cam, 9.35:1 pistons, skimmed and ported heads, inlet and exhaust, 3 angle valve seats, and trimmed valve guides.

Lastly... first time I have actually had a chance to use TunerStudio in anger, if you are still on MegaTune, go get TunerStudio, you won't regret it.

All of the above makes good sense, my WUE graph is pretty identical, I have more or less done the same as Bowie, and although the engine started and nearly stall out it kept running... Tomorrow I will reduce the ASE a bit more and see what happens, I think it'll work, I'll post up my settings if it does

[Hybrid_From_Hell]

Ok

I can help some on this, but I will warn you all now, its a bitch to find and sort. This is becuase there are a number of variables at work here, and you have to be

very systematic and carefull as to what you change, why, record it, and actaully work out whats going on, its like trying to level a crooked 3 legged chair

There are a number of things that are causing this (or could be causing it) I am rather short on time at the moment, so will do my best to spew out enough to try and point

you in the right direction.

Actaul air temp.

We are all in a cold spell, as such this is the underlining cuae, you won't have this problem in mid summer etc, more as to why

this is the case shortly.

On the MS ecu, its not known that 75% of the ECU is to start and get the engine running, and the other 25% actaully then

continues takes over and runs the engine, the fact the engine run means the ECU 25% is sort of on target, 75% and bit and bobs within may need

tweaking, hence why I say it can drive you pottym, theres a lot to alter, and some can make thing wose, also sometimes too BIG a tweak will make it

seem just as bad, when what you have done is go too far the other way and have a differing problem but the same visual effect - if that makes sense.

Also it is liley you will need not a minor change in one place but several minor changes in several places to get the balanced result.

So what are the places ?

Cranking Settings.

These are the initail crank setting, these I would say look okish, and I would leave as these seem to be engine driven rather than engine specific, near enough is normally enough.

Crank + priming table

Rover V8s do need a "Big Slug" to kick them into action, yours is set to 10 - which is a hell of a lot, the ase is after start enrichment, this is a simple "poyur in X amount

of fuel for y secs and then go to ASE table.I would not expect to see 10 in there for a 3.9 more 3-7 max, this over huige plulse + cold = stalling / flooding, once this has happened even in that short time the engine is warmer, which then moves the settings to a different point in various tables....except yours is set to a single number, change stand prime to table and then you get the numbers involves - except ....

the table has a max of 10 !!. On the eales I had

(starting at -40)

26

18

14

13

12

9.8

8

6

5

4 (at 71 degtrees)

This then gives a Differing primes vs temps, yours maxes out at 10, mine at 25 !, your 10 is at mine already was at 12, thats a 20% increase on fuel !

read :

PRIME PUMP WHEN

Values for this item are either Prime Pulse > 0 or Always. If you select the first option, the pump will only be primed if the priming pulsewidth setting is greater than zero. If you select the latter option, Always, the pump will be primed regardless of your priming pulsewidth. You will most likely want to use Always as it will get your fuel pressure up before cranking, but you don't necessarily have to inject any fuel. The priming period is two seconds.

FIRE PRIMING PULSE

The Fire Priming Pulse option allows you to select when a priming pulse is fired. If you select Power Up, a priming pulse will be fired immediately upon turning on the ignition. If you select Twice, one priming pulse will be fired when you first key the ignition on and the second pulse will be fired after a second has passed. If you select After 2 Secs, there will be one priming pulse two seconds after turning on the ignition.

The first option was in the original stock code, however, it will not necessarily work for a cold-start because there there may not already be fuel pressure in the rail when you turn the ignition on. The other two options are just different methods of firing the priming pulse. If you select Twice, then there will most likely be no fuel injected on the first pulse, unless there is already fuel pressure in the fuel rail, and then there sould be fuel injected on the next pulse. If you select the After 2 Secs option, it will give the fuel pump time to build up pressure. With that option, however, remember that if you try and start the car before allowing two seconds to elapse, no priming pulse will ever be injected.

PRIMING PW SOURCE

There are two options for the Priming PW Source. You can either select Standard Prime which will inject a fixed amount of fuel as specified in the Standard Priming PW box, or you can select Priming PW Table which will have your priming PW be based off the Cranking/Priming PW Table. If you would like more control over your priming pulsewidths for varying temperatures, then select Priming PW Table and the values specified in the Cranking/Priming PW Table will be used as the priming pulsewidth for the given temperature.

STANDARD PRIMING PW

If you selected Standard Prime for the Priming PW Source, then this is where you select how long the injectors will stay open when the priming pulse is fired. This value is in milliseconds.

CRANKING PULSE WIDTH TEMPS

Here you can select what temperature lookup method you would like to use for the Cranking/Priming PW Table during cranking. CLT Only uses the coolant temperature and is the original Megasquirt option. This is what most people use and if you have a standard setup with a coolant sensor, this is most likely what you will want to use. If you would like to use the manifold temperature instead, select MAT Only. You can also select CLT, MAT Average which uses an average of the two temperatures.

CRANKING/PRIMING PW TABLE

In older versions of the MegaSquirt firmware, the cranking pulsewidth was handled by a linear interpolation between a high and low temperature pair. A linear curve was not necessarily optimal for some people so now you have the option of setting your cranking pulsewidth at various temperatures.

If you're migrating over from a MegaSquirt version that only used two values, use those as your high and low values in the Cranking/Priming PW Table and then guess the values in between. Then, when you start your car, you can adjust the pulsewidth values for the various temperatures that yield the most reliable and fastest starts.<br>

ASE

A nightmare,.

You have it set to 12 seconds, extend this to up to 20, as this is a diminsing linear curve based on temperature. try this 1st, if not difference over a number of DAYs return it but

I would say 15-18 is better.

ASE then only lasts for X secs - yours is 12 and your table values are F HIGH !

take say 15-20% off those values up to

35 25 22 20 18 14 11 7 5 2 compare 2 to yourrs at 11 !

You may need to play with these between yours and the ones I have just givem why are your ASE so high ?

All are rich, but in effect leaner as it gets warmer, hence the 1st fire and small warmth will then see a shift in table movem,ent

again read

It's been found that during winter the cranking and after start enrichments need tweeking

in order to get the engine to start and run without restarting it several times from cold.

The reason is felt that the after start and cranking enrichments dont follow temperature in a

linear format. So if you find this is the case you can use this table to tweek areas where it

is difficult to start. It has been found that the areas for concern are around 32F (0C)

After Start Enrich Counter is the length of time in Seconds or engine cycles that the

after start enrich runs for. The percentage added is decayed down to zero as the time expires

Set ASE Mode to FIXED for a timed period of ASE that does not decay, after the timer the ASE will

then decay as usual during the TOTAL ASE time period thats left. This has a temp setpoint as

it should only be needed when the coolant temperatures are fairly low.

Setting MAP to a fixed value helps to maintain a constant PW during the initial start where the

map drops quickly from around 100KPa to the tickover KPa. This is held during the FIXED time period. The FIXED period is a period of time at the start of the TOTAL ASE time. FIXED must be less than TOTAL

so, give ase LONGER and lower enrichment, and tweak the problem areas 1st !

WUE

The last on the list

WUE adds a % to the VE table, hence why if you have an engine thats starts fine, a datalog and VE rewite later = starting problem, as the value of the % can be altered, ie a cell that was say 50 and a WUE of "Add 20%) = a value of 50+10 = 60

a rewrite and that cell is rewritten in the VE table to sat 40, now is 40+20% == 48 NOT as was 60, that is a HUGE cahnge !

WUE is best done after the previous areas are sorted, and you have the engine up and not stalling, changing WUE before the above is sorted is going to fudge the issue, have a go with the above and I will add on the WUE at a later date, however, having a PWM does complicate things

NOT having it = a nightmare to get to run as it (engine) doesn't get the lift of RPM + air etc, having it means yet another set of interfering numbers and tables to the mixture. Looking at yours I have a few comments

You can run at 100 or 200, but if 200 vs 100 the other numbers need to be changes. 200 gives more "Detail" frankkly at the moment detail is the last "Extras you wnat so I would just rest to the following :

More reading

Idle control can be used in B&G on/off, Warm up, or Closed-Loop Modes.

--B&G on/off simply opens or closes a valve to provide additional air below a temperature threshold

--Warm up mode linearly varies the duty cycle of a 2 wire solenoid actuated valve to provide additional air during cold startup, while gradually scaling down the duty cycle until the engine is warm.

--Closed-loop mode attempts to keep the engine idling at a set rpm by using PWM with a 2 wire solenoid actuated valve, and actively varying the duty cycle to maintain rpm.

Warmup Settings:

--Lower temp idle frequency is the higher duty cycle required to start and run a cold engine.

--Upper temp idle frequency is the duty cycle the idle valve will be scaled down to as the engine warms up.

These two values are linearly interpolated from one to the other starting from the lower engine temp to the higher engine temp (fast idle temp, and slow idle temp.)

--Idle valve frequency

This is generally something most people will not need to mess with, however, different idle valves are designed to work best at specific operational frequencies. Note that the Frequency of actuation is 1000 / this value

Please Note: true DUTY CYCLE% = Lower or Upper dc * 100 / Idle Valve Frequency value example Freq value - 200 = 50 hz operation, a DC value of 80 = 80*100/200 = 40% true duty cycle

While most valves operate at 100 hertz, some older style 2-wire valves will work better at frequencies as low as 39 hertz. There is an inverse repationship between frequency of operation and control precision because of the algorithm Used to control the PWM output. Lower frequencies have more steps between dc values. 100hz operation has 100 steps and therefore has 1% control steps. 50hz operation has 200 steps and has 0.5% duty cycle steps and that is why the duty cycle values put into MT need to be double the true DC% value.

Some valves may emit a buzzing sound which can be eliminated by increasing the frequency above 100 hertz.

There is a formula for modifying the idle valve frequency. The formula is '10000/desired frequency = x where x is is the variable you input in megatune. For example '100' would be 100 hertz, and '50' would be 200 hertz

Closed Loop Settings:

This is fairly experimental code, but with careful manipulation of variables it should be possible to produce a stable idle.

-- Cranking (dc) is the duty cycle required at cranking.

-- Minimum (dc) is the duty cycle slightly lower than a warm idle dc when regulated.

-- Closed (dc) is the duty cycle where the valve closes.

The values here follow the rule above and are related to teh frequency selected and are not true DC except when the idlefreq values is 100.

Some valves are normally closed (use 0 here) others require a certain DC to clsoe and will regulate above this value. Bosch 2 wire valves are normally open, close at 25% and regulate at about 40%.

--Fast Idle RPM is the idle target for the engine at, or below the fast idle temperature.

--Slow Idle RPM is the idle target for the engine at, or above the slow idle temperature (fully warmed up).

The two RPM set points are linearly interpolated between the two temperatures to provide a smooth transition during warmup

--TPS Threshold is the point above which the idle valve closes in ADC (as it is no longer needed).

--Idle Activation (rpm) is 'how many rpm' above idle that the idle valve returns to operation after the throttle has been opened and closed again. This follows the idle speed as it is interpolated from Fast Idle RPM to Slow Idle RPM

--Dashpot Settle is the wait time required after a dashpot event for the rpms to settle below the idle activation point. After a dashpot it is possible that rpms will overshoot the activation point and this is the time required to settle below activation point.

--Dashpot Adder (dc) is the dashpot duty cycle added to the last controlled idle duty cycle which allows it to recover to a nice idle with minimal overshoot. These values should be low: 1-5 DC

See next help section 'Idle Control (Closed Loop Settings)' for more information.<b

and ...............

More Closed Loop Settings

Deadband range

This is a range of rpm that you do not want it to bother changing duty cycle to regulate idle.

30 to 50 rpm seems to be good here try smaller values unitl it seems stable.

Adaptive idle control

This controls the time between idle control events. The goal is to have a Slow Recovery which is tuned to the engine's time constant at idle. You do not want it to hunt nor have falling idle speed. The Fast Recovery is how fast you want it to recover to a decent idle when the idle speed is higher or lower than the target this can be tuned for startup and to see what is needed to recover if idle speed drops significantly below target for example when the A/C or electric fans kick on and the idle drops below target.

The rpm values determine the curve. The lower value should be close to the dead band value and the upper one should be a bit further away, maybe 100 or 200 rpm

the next 2 values are time constants needed for startup and idle valve closure --Closure speed comtrols the speed at which the idle valve closes. It is the time step delay between each as it steps to a closed dc value --Startup delay is a wait time right after the engine is started that the system stabilizes before it attempts to find an idle speed

However, DO do phyiscal check to see at what number the valve is fully closed, it is a powered close, so disconnect hoses, look inside, switch on ignition and check closed, and allter to whatever number is wants to be to be closed - mine was 25, others are not always the same !

Hope the above helps, as I say took me ages to sort out my same problem

Nige

[Bowie69]

Well this morning arrived, not sure what changed between the first start yesterday and today, but it NEARLY got going and stayed running, but failed. The actual start however was MUCH faster, 3 cranks as above, rather than 5 and stuttering, so that is at least once nugget I can take comfort from -cranking PW and prime seem to be OK.

Current thoughts are.... Firstly as it dies so quickly (<3secs) that I need to do something with ASE, rather than warm up enrichment, looking at my figures they seem around what is recommended, but I will try some more fuel in here, adding around 6 to each of the percentages and try again either late this evening or in the morning, both more likely

Another change I have made to the ASE, it to switch the ASE mode to 'Fixed then decay', and raised the total ASE time to 10 seconds -contrary to what I was trying to acheive, but for test purposes only. 'Fixed then decay' means ASE stays at a fixed value for as long as you specify at the start of the total ASE time(5 secs in my case), and then decays after that. This also allows you to specify at which temperature it uses this fixed ASE, in this case I left it as the standard 24C, above this temp it will decay over 10 seconds linearly. New ASE window:

As we are having problems with the initial cold start, where coolant is under 24C, this will (maybe?) give the engine a little more help staying running after startup, but only at colder temperatures.

Another tool we can play with is the 'MAP Mode during ASE' which can be Normal, where it uses figures from your VE table, or Fixed, where it uses figures from your VE table, but at a fixed MAP value for a given number of seconds -normally you would pick around your idle bins. This is for another day, otherwise we end up in Nige's 3 legged stool scenario

One other thing I did was raise the extra air valve PWM on starting from 50 to 53 (55 when hot was over revving for my liking for too long), and upped the startup delay from 120 to 170 (Idle control - closed loop settings screen) this should help the stalling with some luck.

Will update with results tomorrow.

[Bowie69]

Cheers Nige, As mine actually fires very quickly first thing in the morning, I reckon the crank/prime is OK, and I don't need to alter this to use the priming table, even though there is more control, I don't think I need it...?

[Manxey]

Well I've been trying a few different settings with mixed results.... I think this is just going to take a long time and I'm just not being patient enough!!! I will update my progress if and when it happens

Bowie... Did you have any success?

[Bowie69]

Exactly why I haven't posted either

Drowned it in fuel this morning, plugs plus oven = the cure.... Probably means I had it a bit rich new plugs tomorrow and try again:)

[V8 Freak]

Well as soon as my engine is running I'll be joining in with the experiments.... I also may withhold results pending success

[V8 Freak]

Ok... Not quite able to play yet!! Will be there soon Pete...

Had one of those oh cr@p evenings...

Engine was all back together last night but too late to fire it up and break in the cam.

Have just replaced the cam, followers, rods etc., set the lifters preload, new timing gear, new front cover all as a result of a major oil leak / failure of a thread in the timing cover. (While I was going in far enough to do the timing cover I thought it worth while re-threading two dodgy threads in the block and sorting the knocking cam.)

Finally got to fire it up this evening.

Started and ran like a bag of nails. Rupert who was with me tonight said it sounded like his lawnmower, only worse..

Cut the engine.. Check the basics and assumed it was a change in cam, slight change of timing maybe TPS not right etc. Fired up and sat on 2k RPM for the prescribed 20 mins.. Cr@pping myself all the while because is sounded so bad!!!

When what seemed like a lifetime had finally passed it ticked over pretty good. Drove home in the Eurobox to get laptop and back to the farm to check TPS etc.

Finally found #6 plug lead loose and #7 injector connector not clipped home properly... Much better now its using all 8.

Also found I'd cracked a joint in the expansion tank when I moved the hoses out of the way to work on the engine so can't drive it in anger until it's replaced and the system water-tight again...

Roll on the tank delivery!!!

I will experiment and post up results when I can...

Neil

[Manxey]

Right,

After much messing around with all sorts of settings, ive very roughly tweeked the ASE settings, and guess what, it seems to work, ive been using these settings for a week or so now, and although theres the odd first start where it seems like it'll stall out, she usually starts up beautifully.

Here are my settings at the moment, they still need some fine adjustment, but they seems to be good.

[psoma]

I know this is an old thread, but came across it the other day while trying to diagnose a similar... nay, identical problem with my MSed Rover SD1. It’s funny how particular engines have idiosyncrasies that seem to be fairly common across the make?!?

So the first thing I was working on was cold acceleration (and I mean Australian cold, so 20-40 degree coolant temp). The thing would hesitate badly under a jab of the throttle until warm. Looking at the logs, the MAP would increase before the TPS registered a delta so this thing sucks a lot of air at part throttle. Changed the accel enrichment balance to 50/50 MAF/TPSdot and after some tweaks to the percentages, managed to eliminate the issue. Also noticed enrichment was lasting 2 secs instead of 0.2s leading to an unnecessary dump of petrol so I sorted that too.

Next issue was the one discussed here, and it was harder to diagnose as it didn’t really show in the logs (AFR was fairly steady). Car would catch and fire fine, run for a second or so and then revs would drop, occasionally cutting out. If it didn’t cut out, after 2-3 secs it would sputter to recovery and idle fine. If I turned it off and back on then... not a problem - only the first cold start would exhibit these problems.

Now, I’m not running spark yet as I always wanted to get the fuel dialled in and working first, and idle speed is still “managed” by the extra air valve, so settings around advance and stepper motor control weren’t relevant. This thing started and ran fine on the stock setup with the extra air valve, so I knew it was possible... just wasn’t sure what was causing the problem. Tried messing with the cranking PW which just caused it to crank longer, then last night I bumped the ASE buckets up from 30% to 60% and lo and behold, it started perfectly!

Repeated the test this morning, and again, not even a waver to the idle after catching. Looks like as others have said, these things need a lot of extra fuel just after start. Anyhoo, hope this helps someone trying to diagnose the same.

Cheers,

Tom

[Bowie69]

Thanks Tom, that is a great report, more of that sort of thing helps everyone.

[mad_pete]

I’ll try this with mine. Thanks.