Jump to content

** MegaSquirting a V8 - "Basics of How to" ... The A-Z Saga


Recommended Posts

Note to peeps downloading my fuel map: Since I uploaded it, I've tweaked things a bit and I suggest you do the same;

- Set PWM% limit to 100% (set in error, seems to have no effect)

- Set injector firing to alternating

- Set squirts-per-cycle to 4, this has smoothed things out and got rid of the hesitation on acceleration that some people have mentioned.

I will upload a new map at some point B)

Link to post
Share on other sites
They can go higher than 100V, I'm sure a mate with a Ford V6 has measured 300V from a crank sensor :blink:

Dielectric grease is a double-edged sword, if you get it on the contacts it can prevent them from connecting. I'd stick to vaseline or just a squirt of WD40.

That maybe the case with the sensor open circuit (with only a DVM or a scope attached) as you will get some back-emf, but i'm sure it way lower when plugged into the EDIS. dunno without measuring but i'd be surspised if its over 12v.

landybehr - what you touched was static electricity, I forget the actual numbers but below 1000v you can't even feel it. The same is not true for even very low current electricity, it only needs 60mA to do damage.

Vaseline is what's used in telecomms "bean" connectors and they're used to connect up to 120v in damp atmospheres with pretty good reliability.

C237565-63.jpg

Actually 2-5mA accross the heart can stop it. It also depends on timing i.e if the shock coincides with a particular point in your heartbeat.

is this going OT..... :ph34r:

Link to post
Share on other sites
That maybe the case with the sensor open circuit (with only a DVM or a scope attached) as you will get some back-emf, but i'm sure it way lower when plugged into the EDIS. dunno without measuring but i'd be surspised if its over 12v.

I'm sure the MS VR-input circuit is designed to withstand very high voltages for this reason, IIRC Bruce wired one of the prototypes to the mains and left it for week to test it :lol:

Link to post
Share on other sites
  • 3 weeks later...

Hi,

just managed to have all the MS-specific wiring done. I´m praying for no bugs being implemented by me. One stupid thing happend to the VR sensor (EDIS). I managed to have it tested before installing it and have lost my note which wire colour is + and -; never mind 50/50 chance is fair enough. Nothing worth mentioning (except for a novice like me) but burning the recent code into the ECU worked after setting the baud rate according to the MS-Manual. I have the USB-serial adapter from DIYautotune.

Now that I can play with things a few questions arise:

- about the corrected 3.9-msq file a few posts above. Is it possible to loose some data like the VE/AFR/Ign-tables when playing with MS-Tune offline ? I will soon download it again and save to CD and see what happens.

- When opening a msq-file there is a prompt of "681warning messages written to audit.log.file". Not too reassuring first time you read this. But seems not to influence anything. Best disregarded ?

Now something relevant (real reason to write this):

- I have the 4.2l V8. If I´m correct that equals to 261cu.in (?). When computing the req.fuel I get values of 17.9 which seem to be pretty high. In another msq from a 4.6Disco I read 10.0. But >> the 3.9 injectors flow 54cc´s at 3bar (roughly 43psi) per 1/2minute at 30% duty cycle which sums up to 216cc´s at 3bar per minute. Which value should I trust.

thx, Hendrik

Link to post
Share on other sites

Is it possible to loose some data like the VE/AFR/Ign-tables when playing with MS-Tune offline ?

Well, you can mess it up by putting weird values in and forgetting what you've changed :P but then, of course, you'd save any changes as a new file so you can always go back. ;)

- When opening a msq-file there is a prompt of "681warning messages written to audit.log.file". Not too reassuring first time you read this. But seems not to influence anything. Best disregarded ?

It could be that you're loading an MSQ file that is for a different version of the MegaSquirt software to what your ECU is running. I'd check the MSEFI forums about that to check it's not going to mess things up if you burn it to the ECU.

- I have the 4.2l V8. If I´m correct that equals to 261cu.in (?). When computing the req.fuel I get values of 17.9 which seem to be pretty high. In another msq from a 4.6Disco I read 10.0. But >> the 3.9 injectors flow 54cc´s at 3bar (roughly 43psi) per 1/2minute at 30% duty cycle which sums up to 216cc´s at 3bar per minute. Which value should I trust.

The REQ_FUEL value is fairly arbitrary - yes you can calculate it for your engine, but for example if your fuel map is too lean, you can increase REQ_FUEL and this will have the effect of increasing the fuelling right across the map. At the end of the day it's just a number the ECU uses to calculate the amount of fuel to squirt.

Assuming the 4.6 map works and runs, I'd use that and tune from there.

Link to post
Share on other sites

When I was looking at My Fuel requirement Ian (BBC) had a Fuel map for his 3.5 which seemed to work weel, so the logic was 3.5 vs 4.5 work out the extar Cubic capacity as a percentage and then he upped my MAP on the fuelling to "Compensate"

Seemed logical.

However it ran HUGELY rich, and I with the Megatune and MegalogViewer software reset time and time again, and frankly the VE Map now and fuel settings (on the thread) are only slightly higher than BBCs !

I would personally use a know VE map similar to your engine then go and immediately Megatune and Megalogviewer it with a NB Lamda sensor

I am no expert on this at all - but the above combo nsoftware seems to be :lol:

Nige

Link to post
Share on other sites

Hi there,

(I hope you won´t have to expect too many silly questions from me. Point is the engine is overhauled and not bedded in. I somehow need to have it run (at fast idle) as soon as possible to pamper the camshaft).

I´ve tried to re-load the .msq file for the 3.9 that was posted to be downloaded. Again I only get "0" across the whole VE/AFR/Ignition tables. Is it up to me making any mistake or is the file empty ?

The 4.6-disco´s msq is from a MS-Forum´s guy from SouthAfrica and tuned for high altitude. When I interpret it right it is still used with distributor.

I´d be very! grateful if someone could supply a .msq for a 3.9 or 4.6 V8 with EDIS and Stepper-IAC (the latter optional, I could steal that data from elsewhere).

Link to post
Share on other sites

Landybehr - check further up the thread, I have posted exactly what you ask for:

Second is an MS'n'S-E setup - the engine is 4.6 V8 with "stump puller" cam, running version 029V of the MS'n'S-E code, EDIS-8 ignition

JU_109_46_200708110959a.msq

Here is the .s19 file for the above, it's MS'n'S-E 029V with the correct Rover V8 temperature sender data in it.

Link to post
Share on other sites

Hi Fridge,

I think I know the problem with uploading other people´s msq´s = Unfortunately your 4.6/EDIS-file in MS1-extra and I think that Megatune for MS2-2.88 cannot read it. It doesn´t show any error message and the file msq-info shows yours is 4.0 which is called transferable. Anyway, if yours is the first file of the day to load, every table will show default or "0" (VR/Spark table). If there was any other file loaded before, Megatune will tell me to have loaded your file but keep the data from the file before.

Such a pity. I was really! curious what VE and AFR and spark table you used.

Do you have a WB-EGO sensor ?

Well, isn´t there such a thing as "dump"-mode for Megatune ? I think it will be a great help for me if you could kindly generate such a dump-file and send it to me (Email => landybehr-at-gmx.de)

The other´s 4.6-Edis msq has a spark table based on the Bosch-Thor setup and has a maximum advance of 50deg. This is calmed down a litte by IAT-based spark retard but I´d say a look at a different table will lead me to a compromise.

Thanks !

Hendrik

Link to post
Share on other sites

The connector pins correspond to the coil pairs on the coil packs, so the left pin fires the left towers and the right pin fires the right ones. From that info, and the EDIS-8 wiring, you can connect your plug leads as per Nige's diagram:

coil_wiring.jpg

Link to post
Share on other sites
Well, isn´t there such a thing as "dump"-mode for Megatune ? I think it will be a great help for me if you could kindly generate such a dump-file and send it to me

You can export fuel & spark tables from the tuning screen which are saved as .VEX and can be imported into any version of MS.

To load my MSQ in megatune you'll need to run the configurator (File -> Configurator) and create a new project for MS'n'S-Extra to load my MSQ properly:

config_codev.jpg

Then you can look at the settings and export the fuel/spark tables as required. Don't try to burn my settings to your ECU direct or it will be putting things in the wrong place.

Link to post
Share on other sites
The connector pins correspond to the coil pairs on the coil packs, so the left pin fires the left towers and the right pin fires the right ones. From that info, and the EDIS-8 wiring, you can connect your plug leads as per Nige's diagram:

coil_wiring.jpg

Fridge being the guru on MS that he is - he is 100% right re my piccy above

Now

The Reason the wires to plugs are the way they are is for a minmum of leads crossing over

1 2

3 4

5 6

7 8

The above is the Cylinders on a V8 viewed from above, 1 and 2 are the front of the truck

I basically made up a swappable cardboard templates and payed around with combo until I found what I belived to be the optimum for the least amount of cross overs

What doesn't help is the pic of mine fridge has posted up is viewed from the front of the truck looking backwards, but

1 3 5 & & have no crosses

2 and 4 is minimal 6 & 8 none

then you just swap the outer coil pack wires to get it right for firing :)

HTH

Nige

Link to post
Share on other sites

Just uploaded my fuel & spark tables in VEX format in case anyone's interested:

In the fuel or spark table edit window, use File -> Table Import

The complete MSQ settings file (MS'n'S-E 029v)

Fuel table

Spark table

This is not the new improved version but it works fine.

Link to post
Share on other sites

I have been tweaking and doing a tad more Datalogging when I had the opportunity for a good long series of runs. These were all done (about 3 hours worth in 8 steps) in the LR, and the logs saved each time, but the map constaly updated so I could see what had changed each time when I analysied them later at home

What I have now found is that is the main the cells are about as accurate as I am going to get with a NB Lambda and the MegalogViewer tunning software, when I looked at the 8 logs most changes were on the same cells, and either they were richened a tickle or leaned off a tickle, as could be expected from tunning without the accuracy of the WB Sensor.

BUT

I did notice that there are a range of cells that from all the loggings were not "Goteen into", these are on the chart below, and they are ones I had sussed this on before, but there was a side effect which I will try to explain - which has a knock on effect in bringing the lower LHS bottom part of the map into error....

post-22-1193042848_thumb.jpg

As you will see from the above we are talking about 12 cells which are "Unvisted"

on the 600 rpm 15-40 KPa, 100 rpm 15 and 20 KPa, 1500 rpm 15& 20 KPa and 200-3500 15 KPa

What I have done in the past was just to manually tweak these cells to a value that sort of looked "Right" then the following dawned on me, esp when looking at all 8 logging MAps.

The software takes the cell that it is visting and looks at it, then having worked out as to if too rich or too lean it then takes some of the values from the surrounding cells

ie

123

456

789

Lets say the cells above we are "Tunning" cell 5, it will take values from each of the surrounding cells, 1 2 3 4 6 7 8 & 9.

As the map gets more and more accurate so this compounds the accuracy of the cells around themselves, EXCEPT, if cells 7 8 and 9 are never visted and are set too rich then the 7 8 & 9 "Value" causes the vell 5 in this case to still be too rich, but no amount of tunning will get it right.

If you look at the MAP gain, 1500 RPM and 30 MAP was before tunning higher, and over the 8 it came down around 20+ points to 31, yet the cell below at 1500 rpm and 20 KPa is at 40 !!!, and not only does the value of the cell take bits from around it when logging, it also does it when running as the cells are interpolated !

Thus I decided to hugely lean off the 2Unvisted cells" More, setting them slightly lower than either the cells directly above or to the right of them, and them relogged.

The result is that it immeadiately showed a better low end pick up, the Map has been reset with the pencilled in numbers, and re run.

It is worth doing this, and having the understanding above, this is one of the very few limitatuions of the NB Tunning software, and is also very easy to overcome, just do a number of logs, and copaere each, then adjust those that never get a visit to be in the ame sort of value, then relog again.

Once the exhaust is finsihed I will be going over to WB Tunning, but in the meantime I am frankly amazed how much difference the above tweak has made to it smoothness B)

Nige

Link to post
Share on other sites

Yes, your methodology is correct .............. much the same as my efforts..........however, it shouldn't really make any difference to the low end pick as this is much higher in the table due to the lower vaccum as the load is applied, but ............... in reality, if you are too rich on overrun and you do not have fuel cutoff enabled correctly , then it would leave a fair amount of unburnt fuel in the inlet tracks ............. then it runs rich for a second or to as you open the throttle after an overun. This is further compounded by the acceleration settings (throttle pump) and how quickly you open the throttle. the net affect will be slight hesitation or lumpiness at throttle pick up.

If cells789 are rich and cell 5 gets some visits in a logged run , then cell 5 will be correct as determined by the AFR table, but will only stay correct as long as the celsl 789, 123 , 4 & 6 remain unaltered

The unvisited area is closed throttle and the engine on overrun ………………..

The way I did this was to get all the visited cells correct and then manually enter the values in the unvisited cells (and those cells never get a ‘useful’ visit during normal driving) in a linear manner, taking into account the numbers in the cell above …… i.e if the cell above has a VE corrected number of 41, then the cell below it would be 37 and the one below that would be 34 … etc……

After that excise has been done, then go out and re-log, paying particular emphasis to those cells just above the arbitrary entered values ……. i.e cells 456 in your case………

You will find that it only gets corrected by a very small amount……and the arbitrary values entered will still ‘look’ right, but that is purely academic as these cells only get a visit on closed throttle………… you should be running fuel cutoff on closed throttle to improve the engine braking ……..

It’s identical to a spark table ………….. there are a areas of the table that the engine will never see in anger.

:)

Ian

Link to post
Share on other sites
  • 3 weeks later...

Cracking thread Nige - cant wait to read up on your WB tuning experiences... B)

Quick question of my own. Im planning to go 'serpentine' on the front of my 4.6. Given that i will be using MS when the time comes. Are there any issues i should be aware of regarding crank sensor placement etc. Or will it be just the same as non-serp?

Link to post
Share on other sites
Cracking thread Nige - cant wait to read up on your WB tuning experiences... B)

Quick question of my own. Im planning to go 'serpentine' on the front of my 4.6. Given that i will be using MS when the time comes. Are there any issues i should be aware of regarding crank sensor placement etc. Or will it be just the same as non-serp?

Glad the threads of use and interest B)

erm

Why esp are you thinking of serpentine ?

On Mr fridges motor wot I shoved together for him built so carefully and painstakingly :P, he had a sepr crank but with a 3.9 efi non serp front mit a spacer on the crank end, this way used all std 3.9 non serp bits ?

Just cos is a serp doesn't mean you have to go serp route ?

WB tunning is coming - F Big Bore exhuast 1st :)

Nige

Link to post
Share on other sites

The reason for wanting to go serpentine is simple. i want to use a f huge alternator without the poxy non-serp belt slipping all the time!! :lol: and i dont want to use 2 alternators. going serp will keep it neat and simple....hopefully :lol:

Link to post
Share on other sites

I also heard the serp oil pump is meant to give better flow at low rpm.

The 4.6 block has it's own crank sensor at the rear of the engine pointing towards the flywheel (you'd need a 4.6 flywheel too which has drilled holes for tooth gaps), i think this may be a better way to go as it's more protected from offroad debris, etc. I don't know if this can be used with MSnS, anyone tried it?

If you were to mount the triggerwheel on the front of the serpentine, i'd have a good look for possible mounting locations / bolts for the VR sensor bracket to fit onto before going down this route.

The non-serp frontend has a couple of nicely positioned bolts on a flange that look almost made for the job. On mine (4.6 non-serp) I made an ali plate which bolts to the underside of the front of the chassis to stop anything flying up and hitting the sensor, doesn't seem to affect cooling and keeps a lot of cack out of the engine bay when charging through deep muddy pools... something you may find yourself doing a lot more with MSnS :lol:

Also, is anyone running or has upgraded to MS2/Extra? Interesting stuff going on at MS2/Extra Development Forum with EAE, idle control and customisable outputs.

Link to post
Share on other sites
If you were to mount the triggerwheel on the front of the serpentine, i'd have a good look for possible mounting locations / bolts for the VR sensor bracket to fit onto before going down this route.

I know, but the thing with that is i need to get hold of serpentine parts (ie crank pulleys, timing cover etc) to find out if i can do it. By that time ive spent a fair chunk of moolah....

I was just wonderin if anyone has megasquirted a serpentine engine that could tell me its ok, or not as the case may be... ;)

Link to post
Share on other sites
The 4.6 block has it's own crank sensor at the rear of the engine pointing towards the flywheel (you'd need a 4.6 flywheel too which has drilled holes for tooth gaps), i think this may be a better way to go as it's more protected from offroad debris, etc. I don't know if this can be used with MSnS, anyone tried it?

I did think this through, whats against it is any probs with wheel / sensor means engine out, and a lump of 6mm Plate thin repair metal :P plus chunky bracket :P :P means fairly indestructable but easy to get to :)

If I sharpened the teeth to a point I could then also have a powered chipper 36:1 saw for when stuck on a stump :lol:

Nige

Link to post
Share on other sites

The 4.6 flywheel is 60-2 if memory serves, so you can't use EDIS but you can use MS'n'S with direct coil drive, as Rog (Istruggle2gate11) has done. It's more faff to set up but does give you a couple of extra toys - spark cut rev limiting being #1.

I did think this through, whats against it is any probs with wheel / sensor means engine out

If something damages the flywheel you've got bigger problems :blink: the sensor just mounts through the back of the block (like the starter motor on the other side) so is easy enough to get to.

Link to post
Share on other sites
  • 2 weeks later...

Oh DOH - permission to say "B*gger" :lol:

my "Blinging" effort today found me a mistake :huh:

Today I fitted the Eales V8 Rocker Covers .............

sort of been meaning to do this now for a while - layers and layers and layers of paint off .......and a clean up finally saw them ready.

I had used a Volvo Extra Air Valve thingy, except I had never wired it up......

I had decided that I was going to remove it completely, made a plate for the rear of the plenum, and junked the lot,

This EAV had come off a 'scrapper' and had just been shoved in with 2 hoses.

Made the plate 3mm ali, drilled and scred on, ....then made a bung for the hose on the side of the plenum, .......then saw the EAV out of the corner of my eye. :(

post-22-1195313520_thumb.jpg

Ooops !................When I had fitted it I had just done that - didn't check - it was open :o

Should have been closed, ........obviously stayed open on the scrapper when moved in the yard, stayed that way as I never rewired it up ! !

Anyway a quick data log and drive about saw the map redone - it leaned off at the lower end, and increased at the top end,

A chat with BBC as to the why and wherefores confirmed my thoughts, rather than explain the issues here - the simpler way is to say - if you fit one then either wire it up or ........

.......ensure its F closed if not :lol:

Anyway some bling - for those V8 fans see Page 175 of tunning Rover V8 Engines - by David Hardcastle B) ring any bells ?

post-22-1195313455_thumb.jpg post-22-1195313503_thumb.jpg

post-22-1195313650_thumb.jpgpost-22-1195313669_thumb.jpg

Yooo Bling innit :lol::P

This means the old ones that were on it are up for sale :

post-22-1195313737_thumb.jpg

These are a set of V8 Development Rocker covers, either for those who wish to Bling their Mota B) or as for the real purpose of these they are made to take the taller rocker shaft asemblies as with the Eales,

They are machined to take Oil filler / flame trap, and breather on passeger rear rocker - which I have made to take a washer tube size tube that then connects to the breather unit relocated high up :) either way bare ali nice condition, can either be collected of I can post - £100 pair collected £110 posted, Paypal or cheque / cash 1 week if not sold onto e-bay. These could as I have done above with the 'JEM' Ones paint in self etch (halfords) spray with Black gloss, then lump of plate steel, and wet and dry, water and then rub off the top paint etc to get back to the ali ...

Nige

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

We use cookies to ensure you get the best experience. By using our website you agree to our Cookie Policy