Fatboy

Nobody got any thoughts or advice?

The Lambda sensor in the kit is a narrow band. I've not seen any suggestion that it can be converted to a wideband by twisting a couple of wires together, I think you've maybe misunderstood something there.. There are two whites for the sensor heater (A supply and ground), a blue signal wire to the ECU and a sensor ground. I initially joined one of the white wires to the sensor ground wire and took that back to the relay board but then read a post somewhere that said that those two grounds should be kept separate with only the sensor ground being taken back to the ECU to try and avoid "noise" disrupting the signal from the Lambda. It recommended earthing the heater (white) to the chassis away from the ECU. I did that. For the EAV valve, there should be a setting where the EAV is totally closed, it will be "closed" with a signal voltage, you just need to figure out what that voyage is. The reason it does not close fully under "no" voltage is so that if it fails, it will still be a little bit open and allow the car to idle. If its the Bosch valve, then my story here may help. Disclaimer: I may be talking rubbish so take anything I say with a pinch of salt. (I'd be more confident if mine was running properly... )

NB sensor in, map from a couple of weeks ago loaded, EGO step set to 0, runs quite well... (EGO set to zero should mean it won't self-adjust...?) I am wondering if I've screwed up the AFR table and thats whats causing the deterioration when I try autotune... and my lack of familiarity with it all has me in the dark somewhat. I've googled and read the manual, but not found any clear guidance. So... when I was trying to narrow band tune, me EGO Control settings were: With the NB, and Controller Step Size set to 1, the VE Table figures change which I assume means it is trying to home on on AFR targets from a reference. Indeed, on the VE Table, if I double click on a cell, it tells me how many times that cell has been hit, and what its target AFR is. The bit I can't figure out is, where is it getting the AFR figure from, because I had the AFR Target Tables' boxes both off: Then, if I do turn the "For VE Table 1" box on, and view the AFR targets, I get: The AFR Table displayed in VOLTS! ..... Is there some way that I need to convert my target AFR's to volts? This is where I think I may have made the error before, I might have input targets without noticing the table was in volts, thus was asking it to look for 14.7V etc, which might be why it tipped the fuel in.. Thanks for any tips / pointers you can give.

Dave, Thank you. I went back to the NB and the last good map, which was with the NB. I had checked settings in the settings, and in the EGO, all okay. Very good point re the Lambda measuring O2, I hadn't twigged that! Air contains 20% oxygen so it's probably okay....

That post above should have said "I'm only tuning above 1200rpm".. I loaded a map that had been working and tried again, the VE figures were going mental and it did same again, loads of fuel and had to pull over, reload the map again and abort tuning to get home. I tried moving the lambda earth from the redundant Fast idle relay terminal to the main Earth on the relay board, no difference, still 0.2v with no engine. My amateur theory was that if it is wrong, the 0.2v could thrown the tuning out..

Gents, I.m really struggling with this... it seems to be going great for the first few minutes and then whilst autotuning at "cruise" speeds, throws a wobbly... then runs like a pile of junk. I'm only tuning AFR about 1000rpm. I've gone back to the original NB lambda, because it had been doing okay on that, now it does the same with it. Something that has been bothering me - What should the lambda voltage be with the Megasquirt powered but the engine off? I'm getting 0.2V on the NB and can't see where to calibrate it.

Ah.. I'll adjust the AFR table then and try that. Q: but how do I sort the range on the Lambda gauge?

I fitted the Spartan2 wideband and adjusted the settings with some gladly received guidance from Quagmire.. While fitting the wideband, I split the earth cables from the sensor and earthed the heater -ve to the chassis, leaving the sensor -ve going back to the relay board. (As recommended in the literature). Fitted the wideband. In the Project Properties / settings / tab, changed the EGO O2 Sensor to the Innovate PLX 0-5V 10-20:1 AFR. (Same setting as Spartan 2). Then, on the main Tunerstudio screen, went to Basic settings and changed the sensor type from Narrowband to Wideband. I also changed the switch point. Then set the following AFR target table (copied from one of Bullbar Cowboys posts on here) Then took it for a spin.... The starting point was the last tune from the Narrowband: and it did start to make some changes in every cell it hit... but weirdly, the Lambda sensor gauge stayed red, on the 0 to 1v scale for the Narrowband. I checked the settings and they seem right so assumed maybe it needed to calibrate.. However, it was changing cells and the volts from the Lambda seemed okay.. After +/-20mins, it started to hunt at low revs, and didn't seem very healthy but at more than half throttle, it was fine. The final wideband table was: I had wired up the LED from the Spartan2 which blinks when it isn't at the right temperature and it was only blinking for a few seconds after the engine started. The sensor is at a point in the exhaust beyond the Y piece so getting gas from all cylinders.. Again, I'm baffled. I've checked and rechecked it but just can't figure it out. Any tips will be sincerely appreciated. Thanks for reading this far.

Many moons ago, I had a random alarm problem on my '96 Discovery. It turned out that the plunger had got broken off the bonnet position sensor. IIRC, it was at the rear of the bonnet and was supposed to be depressed when the connect was shut. With a chunk broken off, I guess it was at the brink of making contact so occasionally would trigger the alarm. Easy thing to check.

Doug, Before having a conversation, taking everything into account, see if you can figure out what value you would be happy to pay. If you can, estimate the costs in categories (pets, mechanic time, MOT, rolling road time etc) Then, during the conversation, it gives you a chance to make an offer and pay what you believe is a fair amount. You might even consider writing a letter to explain, and send them a cheque. Legally I think this prevents all sorts of "non payment" action and you may find the problem can be resolved to both parties satisfaction.

Oh god... Now I'm on the Difflock website. They also do various mirrors but details are vague and I've no personal experience. "Hmm... drain plug"

Fantastic Chris... thanks for that. Me: "I'm sure Mudstuff do mirrors, I'll check..." Browse to Mudstuff website to check them out, and they indeed do larger mirror heads. However, if you already smash standards, you'll definitely break these bigger ones... Maybe not for you. "Ooooh... While I'm here, I'll get some cargo nets for the 110, and some fixings.." "LED outside lights.. I like those...."

Pete, While mine was out I gave it a really good clean, fitted new end plates from YRM because the old ones were a bit scruffy and corroded, then spend a bit of time sticking Flashband on the underside of it. Once it is in place, I go over it lightly with a hot air gun which softens the adhesive. I use one of those narrow (2" wide?) wall paper rollers to really get it moulded into corners and any changes in profile. I also stick it onto the top surfaces, cutting out the holes for seat / handbrake bolts using a piece of copper pipe as a die. Looks quite smart in a gunmetal grey kind of way... it must make some difference to the noise but I can't really tell to be honest... I fitted a marine battery isolator on the front of the box (For 2 batteries - 1, 2, 1+2, off), an Anderson 175 Plug and the little tail that comes with a Ctek charger because I've bolted Mazda seats in and it takes a bit of effort to get to the batteries. YRM also do moulded corner pieces for the corners at the front of the box that get kicked on the way in/out. Excellent for hiding the edges if you fit carpet or something. I initially thought the fit could have been a bit better until I realised I'd fitted the wrong on, correct it and they are an excellent snug fit.

Jim, I made a Heath Robinson solution at the weeken that was actually quite successful.. I knocked the ceramic centre out of a spark plug, then glued a short piece of pipe (short enough for the plug socket to fit over). Luckily, I had a long bit of see through hose that is a snug fit onto the pipe, popped the other end into a plastic bottle half filled with oil. Clip the plastic pipe to the raised bonnet so there is a good hump between the bottle and bore. Key bit - make sure the bottle can vent air out. (I didn't on my first attempt and pressure tested the system...) As you turn the crank pulley with a socket, the pipe will blow bubbles through the oil, then as it reaches TDC, the bubbles slow down. As soon as the piston starts to descend, it wants to suck oil out of the bottle. Sounds rough, but if you watch the behaviour of the oil, you'll get a good, accurate point. Use oil just in case some gets sucked too far up the pipe and heads into the bore.

So.... accessed the relay board and was grateful I'd labelled the group of cables from the Lambda sensor. I remember reading something saying that the ground cable from the Lambda needed to be earthed near the ECU and because there were no spare points on the board, I'd stuck a male spade terminal on it and used the Earth terminal on the spare relay plug... except I hadn't, I'd stuck it into one of the terminals that is live. Pulled it, connected it to the proper negative point and started it up. Voila! Lambda stayed below 1V and got the first time ever, auto tune actually started to work. Took it for a quick run and can see a difference already, might even auto tune it again tomorrow on the way to work! ? Thank you all for the help, can't believe it was so simple.

Oh no worries, and i wasn't being a smarty pants. I just listed the correct one in case someone else is looking.

Indeed, black on Lambda goes to blue on loom to ECU. Grey on Lambda should go to black in loom to earth. I suspect I've gone wrong back at the ECU somewhere.

Thanks for the pointers everyone. Bit of an update - I had a quick check tonight.. The gauge shows 0.2v if I disconnect the sensor from the loom. The black wire from the loom (should be earth) is providing 12v... I've written a note on the wiring diagram that it should have been earthed at the ECU so I suspect I got it mixed up with the White/purple trace wire that should T'd into the supply to the fuel pump (it is showing earthed). I'll check the other ends of these wires at the ECU after work tomorrow. Had a look for a Spartan2, the 14point.com website doesn't exist but 14point7.com does. Nearest dealer in Holland so I've ordered one. ? Promise I'll keep this updated as I make progress, thank you for the participation!

Indeed, narrowband...

Morning gents, Please bear with me on this post as I’ll try and provide as much information as I can think of. If you have suggestions or are able to help, let me know - any assistance would be much appreciated as I’ve been struggling with finishing this build off for an embarassingly silly amount of time… Engine is a 4.6 V8 with tophat liners. Fully rebuilt and Megasquirt installed (in 2013, ahem…) MS-1 v3, 029v. Loaded with Nigels 4600 Base Program. Injectors are new Bosch 0280155900. Supposed to be a good solution and used by Ford in 4litre, 6 cylinder engines. Req Fuel is 19.4 (4600 displacement, 8 cyl, Injector Flow 200, AFR 14.7) Plugs are NGK BPR6ES Lambda Sensor is a Fuel Parts 86400 narrow band, 4 wire Zirconia. (From Nige, with the kit). Timing has a trim angle of 5 degrees. I checked TDC with a hose connected to an old plug shell, run into a bottle of oil so I’m confident it is correct. Then figured out the trim requirements following Fridges recent response in another post. Throttle Position low is 12, high is 208. PWM valve is a Bosch unit. Leads are ex-Nige Magnecor variety. I subscribed to TunerStudio at the start because I'm not very confident with this type of stuff and wanted to develop a familiarity with the programme I wanted to use long term. This seemed to be the favourite, so I went with it. Over recent months I’ve changed plugs, leads, injectors, Lambda sensor etc. but just cannot get it to run properly... It starts fine, ticks over okay. Warms up and drives but the O2 sensor goes out of range therefore it won’t autotune. Tonight, ticking over, the O2 was at 1.4V but as soon as I take it down the road it climbs to 4.5v… When I take the plugs out they are black so seems to be running rich… I took the photo below to show the typical gauges when stationary. From there, the O2 only get worse when it is running..

If perchance you are in the Aberdeen area at all, I can lend you a set of four modulars with MTR's on them. (I don't need them again until next winter...)

Thanks Bowie... Time for a little update - By peering at it, I was convinced that the "gate" in the idle valve was not full width (see photo above) but hooked it up to the loom a couple of nights to try anyway. I messed around with the frequency settings as suggested with no luck, then did some googling to see if anyone else had figured it out. TomG (on here) had done some trials in May 2007 but he had a Bosch IAC valve (0280 140 505) and needed to install a diode to bring it alive. My shoulders dropped a bit when I saw that, although I have a Bosch 0280 140 516 but I thought I'd try the settings he used anyway (without the diode). Valve Frequency 200 Crank 90 Minimum 70 Closed 60. Amazingly, looked in the valve and it was totally shut. I figured that that was probably because I'd had the engine running so it was warm enough, so would have to wait for it to be cold to see if it opened... Last night, rather than pull the hose off to see if it was open while cold, I flicked the key and she burst into life, then happily sat idling at 1000rpm. Bingo. MOT is now booked (again) to see if I can get a "pass" that will let me take it out for tuning runs... Thanks again!

Morning "wiser ones than I", I'm foraging forth in my aim to be "the longest time to complete a Megasquirt installation"... I'm currently on another push to finish my 110 and as it is a 4.6 plenum, needed to install an idle valve. One of my errors, (which I realised recently) was that I had removed the stock idle valve, blanked off the entire ID with a snug length of hose, then capped it with a little alloy plate. This accidentally blanked off the hex idle screw.... I've now got it idling at about 1000rpm with the hex screw and adjusted the stop under the throttle lever. It does not like going lower than that at the moment. I've got a 2 wire Bosch idle valve (from Nige) which i believe is out of a Volvo? I tested it with 12V and indeed, the little rotor rotates to expose more area but when there is no power to it, there is still quite bit of throughput.... When you look down the outlet, you can see the shutter is not the full width of the outlet. Is this normal? If so, I presume it is normal then to have to reset the plenum screws to lower the idle again once this valve introduces its air contribution to the inlet, even when "off".

Oh god.... Me too.

I applied one coat, it would easily obscure brush marks... Rob - perish the thought of rubbing it down if I needed to cover it. The tintable variety is translucent. You tint it by adding truck paint - easy to do.