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

[Bowie69]

Yup, perfect sense Nige

So it is almost akin to having noise on the TPS channel?

My TPS value is 28 BTW, just to make it look odder I think I set it to 33 for PWM to kick in, mayb be worth me trying higher I guess...?

One setting that I found can interfere is the fuel cut off on overrun, leave this switched off when tuning PWM, especially if you are using a low figure like 1300rpm, as it will interfere with it badly, shutting the fuel off when it is trying to drop the revs, a much more sensible figure is 1600RPM in my experience, but only once you have PWM working as it should.

[disconc]

Hi Boys,

thank you for your feedback. Every bit of information gives food for though on this matter

Cheers

Erik.

[tarmacshredder]

Hi all,

quick update as I haven't posted for a while as I haven't got any further as I have parked the land rover up and ignored it as I had got fed up of hitting my head on a wall.

I did get some help from Zoltan who put a wide band lambda on and we tuned the idle so that we were somewhere near and not massively over fueling, with his help we also moved the ignition advance so that when trying to pull away it didn't stall everytime and now will pickup and move off without any issues.

It was used last in the snow and was nice to use it again we got maybe 9mpg max which was crippling and the smell of fuel burnt your throat and stung your eyes after a bit but it didn't get stuck and although the starter was a little lazy at below -10 it started with the 3 attempts which I have listed before with out fail. The last day of the snow I cracked the screen on a snow laded branch that was stuck out of the hedge and it hasn't moved since.

I am looking for some local help, is there no one in the warwickshire/worcestershire area that can give me a hand and try and workout whats up as I am fedup with it.

Glenn

[tarmacshredder]

As you can see no reply to my post, does no one in the midlands run megasquirt on there Land Rovers or Range Rovers?

I have money here to pay someone if they can sort it out.

Glenn

[Bowie69]

Well, TBH, you gave very little more information, but I took a look at your log file anyways, though I think most points will alreayd have been covered.

Firstly, have you sorted out the TPS issue? It's peaking at just 34, which is no where near high enough, this means one of two things, one you haven't claibrated it, or two (which in this case I think more likely) the TPS has failed. This is probably why you are getting a drop in RPM when you press the throttle, as there is no/very little acceleration enrichment happening.

MAP readings seem fine, 30 is OK for idle.

Have you checked to see that the spark advance is changing with RPM/load with a timing light?

Turn fuel cut off on overrun for the moment, you could just be confusing the issue.

In all honesty I think you probably need to do a little more reading of the Megamanual and tuning techniques, but the very first thing you should do is go back and verify ALL your connections to sensors, ensure that the wiring is of satisfactory quality and no loose connections. Unplug the loom from the MS and start measuring resistances, testing components like the TPS, CTS, MAT, triple check your HT lead order, then test you have expected voltages at the EDIS (there's something on the internet somewhere about troubleshooting these, AND get your lambda connected and configured in MT/TS properly.

No point trying to diagnose a running problem if the loom has errors and the software isn't setup correctly.

I guess having got to this point you are running Nige's MSQ? If not, do.

[tarmacshredder]

Hi Bowie69 and thanks for the reply.

I started the Landy this weekend for the first time since the snow, turned key and fired up straight away, it stumbles, snorts and lumps around for about 20 seconds and then ran at idle. All eight cylinders are running fine,you can blip the throttle and revs freely.

TPS was set again since the log and now appears to be giving values similar to what I have read on the forum, so I dont think thats a problem.

msq is one that nige put on.

My big problem is I cannot tune with the lambda as I have a residual voltage (shown on megatune) that fluctuates even with the lambda sensor disconnected from the ecu. When the lambda sensor is connected to the ecu I get the same residual voltage.

I have had the loom off the landy on about 6 different occasions and have pinned everything out and checked resistance for good connections and no cross feeding. I have sensor values on megatune that I would expect and look quite good, but then this is the problem, I have looked and looked and tested and tested and I am still no further forward than I was 12 months ago.

Zoltan was able to put a wideband lambda on and tune the idle as it was massively rich, we were also able to adjust the spark advance at the bottom of the rev range which cured the pulling way problems that i was experiencing.

I haven't got anyone to help as the person that was helping hasn't helped for over 12 months and has run away as they too cannot see whats up.

I have run it with a timming gun in the past but when your on your own you run out of hands to test and watch everything at the sametime, this is why I am desperate to find someone that can help me locally as I dare not run the landy and distance due to fear of damaging it.

Glenn

[Bowie69]

Fix the residual voltage on the Lambda, there must be a reason for it, then you can tune. If it's a problem with the ECU, read the schematic and work out why, and if there's a duff component on the board. If you're not up to this then it will need to be sent away for repair.... your original purchase point might be good.

If you want someone to try it on their truck to see if the residual voltage appears for them as well, I am sure there will be a couple of volunteers (me included), but it will be a post back/forward job rather than me visiting you

I reckon now you have set your TPs properly, it's time for another MLV file, email me it at peter.bowater(at)zen.co.uk and I'll host it. If you can, take it for a drive whilst logging, will give more useful information.

Other thing to do is unplug the PWM idle valve, it will default to an open setting, the stalling after start could just be the engine not getting enough air, or it could be your after start enrichment settings, this will help rule it out a little.

[tarmacshredder]

It is not for a want of trying to workout why there is a residual voltage on the lambda but I have a V3.57 board which is surface mounted and I cannot workout why I have a problem with the circuit. I am pritty certain the fault occured after the mod for Edis8. The mod for Edis8 isn't in the megasquirt manual for the V3.57 board so I got the information from Diyautotune in the sates that sells these surface mount ecus and they carry out the mod for customers there. I have emailed them with the problem I have but they didn't have a solution to the residual voltage on the lambda circuit. The company that supplied the ecu is no longer trading and it is well over a year old.

My request for someone local that has a knowledge on megasquirt, is so I can do exactly that swap over the ecu and see if there vehicle has the same residual current but also for them to have a look and hopefully see something stupid that I have done that is so blatantly obvious to them but I cannot see.

I like your suggestion of unpluging the pwm valve and see if it makes any difference when starting as I haven't tried that and no one has suggested it before.

Glenn

[TheRecklessEngineer]

Ok, reading back through your problems again, it seems like everything is sorted bar the lambda problem.

The first thing you need to do is to disconnect the lambda (leave ground and heater wires connected) and stick a multimeter on the signal wire/ground. If you get the same 'residual' voltage, then your lambda is the problem. If not, then the ECU/wiring is at fault.

[Bowie69]

As above you really need to sort the lambda residual voltage issue, and first step is to determine whether it is ECU or wiring at fault -I think from what you said it is ECU, in which case it's time to talk nicely to the people that did the EDIS mods.

Then you can go tuning! In all honesty you don't need it to be that much too rich to start burning and stinking, from what you say you are 95% of the way there, keep at it!

If you did want me to plug it into my truck, then by all means post it to me and I'll send it back after testing, just can't justify that kinda round trip mileage

[TheRecklessEngineer]

I'm happy to take a look at your ECU - but I can't spare the time just at the moment. I might be able to do something in a few weeks though.

[FridgeFreezer]

Have you tried a different ECU modded the same? If you talk nicely to Nige he may have a spare unit he can send you, but you will need to be 100% sure the EDIS mods are wired out the same way.

I may be able to have a poke at your ECU by return of post, but it won't be quick as I'm flat out busy at the moment.

Fundamentally your problem is it's running pig rich, which means barring injector/major ECU fault, it's set too rich either in REQ_FUEL or in the fuel map. If the lambda sensor is the only fault then disable the O2 correction, imagine it's a carburettor and tune it by ear/nose/seat of pants.

[tarmacshredder]

Unfortunately I haven't got access to another ecu so cannot try that to see if the problem is related to the vehicle or the ecu. (Again my plea for someone local to give me a hand). The db37 plug is wired and setup the same as in the megasquirt manual for a MS1 version3 so that it can be swapped and interchanged onto another vehicle, I also have a db15 plug that has the additional mods wired to (fan switching, dual table switching and the original direct coil drivers, non of which are connected).

As for tuning without the lambda/megatune I am nervous, it is obvious that it is over fuelling at the moment due to the smell and consumption etc. If I drop the required fuelling overall it runs rough and starts to lump about, I can drop the individual cells on the fuel table and make no noticeable difference to the engine note but I am conscious that I don't want to go the other way and be running lean as the 4.6 is then going to have all kinds of other problems that are going to be even more expensive to fix. I can try with the multimeter attached to the lambda but I am only going to get it to hit certain cells and then there is going to be a lot of thumb in the dark calculations to try and get the fuelling table to look something like, and the whole idea of the megatune is so that you can log and tune accurately so that the ecu is matched perfectly to your engine.

Seams odd that I appear to be the only person in Worcestershire and Warwickshire that runs megasquirt on there Rover V8! I am not sure if that makes me feel privileged or depressed! lol

Glenn

[FridgeFreezer]

Glenn,

Talk of all the various mods to your ECU make me uneasy, any chance you can pop the lid and take a decent photo of the insides & post it up? Just that I've seen some proper abortions, even done by "professionals".

[bobtail84]

Glen, this may be a stupid question and please tell me if you have been asked, but is the MAP pipe connected to the ECU and the plenum? It must be on the engine side not pre butter fly.

Jeff

[tarmacshredder]

I have taken some pictures of the board, the wires which run out of sight, the large black ones and the thin purple, pink, brown and grey are all to do with the coil drivers which arent being used.

MAP sensor connected to ECU and back of plenium, have had this disconnected before and it was alot worse then! lol

Glenn

[Hybrid_From_Hell]

all to do with the coil drivers

Ah ha !

My advice - change to edis 8 an bin direct drivers - pita to sort, many have tried failed an gone edis

Nige

[tarmacshredder]

It is now runing edis8, "all to do with the coil drivers which arent being used" they are just still in the box switched off and appropriate jumpers moved for Edis.

Glenn

[mad_pete]

I've run my MS auto with and without the idle valve. I'm playing with it at the moment but the PWM does run away a little bit. Mine worked fine without such a valve at the moment I'm going to try dropping the values a little bit as I'm forever on the brakes to stop it running off.

Mine runs at 1200 rpm in neutral but drops to about 900 in drive.

I'll see if I can dig out the values I'm running over the weekend.

[FridgeFreezer]

Glenn,

That doesn't look brilliant - photo is very small, any chance of a more detailed one? Where are those wires going to - stuff bolted to the lid or what?

[tarmacshredder]

Yes there are 4 coil drivers an inline fuse and an earth all on the lid, I will try again to take some more pictures but the components are so small it is very difficult to focus.

Glenn

[elbekko]

Try using the macro mode on your camera (the little flower-labelled button).

[tarmacshredder]

Progress, having had he ecu off to picture the inside, I have once again checked the wiring for the lambda sensor and traced it back to the ecu. I have reconnected the lambda sensor wiring to the db37 plug, put the ecu back in and then fired it up. Voltage on the lambda was showing 0.20volts before start up and after 40seconds or so the voltage rose to 0.69-0.71volts. Thats the good news, I am guessing a bad connection in the lambda plug but not really sure why it is now working and wasn't before. I still have the residual voltage and what I don't know is wether the 0.69-0.71volts is correct or whether that is ontop of the 0.20volts I had before. Also it is now not running on all cylinders having lost 1 or 2 on the drivers side bank I think, it is popping and fluttering in the exhaust header on the drivers side, the lambda is on the passengerside bank.

All good fun!

Glenn

[AncientGeek]

Serpentine notes.

Adding some notes after a successful serpentine install, '97 D1, Edis8.

I built my own ECU from a kit, be aware that you need quite a bit of rtfm and a few extra parts that are not in the kit are needed.

I assembled a stimulator first and this gave some good soldering practice and also came in useful later.

My pulley had a UFO shaped dirt trap with welded nuts. I had to bin that and add some nuts in order to fit the triggerwheel and spacer without fouling anything.

If I were to do the wiring again, I would leave the existing loom in place and add a small extra loom for the extra sensors and coils. I still have nightmares about fighting with the spider wiring.

I used the two existing Bosch 5 pin relays. The setup required differs from the kit-supplied and commonly available wiring diagrams which are for older models.

Main (Load) relay:

Pin  Was						Now


30   Brown-green (fuse box)		  Same
85   CPU/12 (Blue-red)				  Earth
86   Brown-Orange(Fuse box)		 White-Green (Ign Switched)
87   Fuel Inj+(Brown-Orange)		Same
87a  Brown-Orange(ECU-2)		   Blue-Yellow (MS Power) + EDIS8 Power+Idle Valve

Fuel pump relay

Pin Was							 Now


30   White-Purple (fuse box)		 Same
85   Blue-Purple (ECU)				 Blue-Purple (MS)
86   Ign Switched (white-green)		Same
87   Black (Fuel Pump/Spider)		 White-Purple (Fuel Pump)
   White-Purple (Purge)			Not needed
87a  White-Orange (O2)			  White-Purple (O2 heater +Coil Pack power)

TPS

The wiring colours on the serpentine are different to earlier models. Wire as follows:

ECU ---- TPS

Yellow ---- Brown
Green ---- Green
Red ---- Blue

HFH's coilpack mounts are a work of art. I had to use a b*st*rd file to take a few mm off the alternator casing otherwise it was hard up against the rear coil. I put some rubber around the top bracket as it looked like it would chafe the upper radiator hose. I also had to trim the upper radiator clip on cover with a dremel to clear the coil pack.

I fiited my ECU and EDIS8 in the swing down fusebox door.

Fixing the Aircon

The aircon wiring depends on the Lucas ECU and stops working when you disconnect and remove the old ECU. There is an easy solution. Remove the yellow-black wire from the AC logic relay, and put it on pin 86 of the AC clutch relay. You won't need the AC logic relay or the fan timer unit any more.

It might be an idea to configure one of the extra outputs from MS to an “engine is running” signal. Then wire that through the AC logic relay so that the AC clutch can only be engaged when the engine is running.

MAF Replacement.

I found some 76mm OD aluminium pipe (80mm would have been better) and cut it to size. I found a silicon 76 mm to 62 mm converter at a Chinese shop, cut it halfway along the downsizing , and siliconed it to the air filter outlet. The IAT is fitted to the air filter housing. Heatsoak is still a problem.

[travm]

Since this is turning into a bit of an MS-tech thread, here are a couple of maps for people:

First off is a standard (B&G standard code, 8x8 tables, fuel only) settings file from my RR 3.5 V8 after a bit of datalogging & tuning, the idle isn't perfect but the rest of it is close and I have been running these settings (albeit with a less well tuned map) for years now:

JU_RR_35_200709062328a.msq

I would use this as a base settings for any stock RV8, just up the fuelling (REQ_FUEL in constants) by a certain % for larger engines.

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, again this has been datalogged & tuned up and is actually pretty good, certainly I don't feel the need to tweak it any further right now (although if Bill & Dave turn up next weekend it may get fiddled with):

JU_109_46_200708110959a.msq

Edited to add:

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.

Hi guys,

Assuming you're still around - but the links to these files are no longer...I'm trying to find some base to start with a tune on a 4.6 stump puller cam MS2Extra but everything I can find on the interwebs is invalid when it comes to VEX or MSQ files.