Adamw

Ignition does (wasted spark), but fuel doesnt.

The ECU needs returning for modification or replacement. RMA = Return Material Authorisation. We dont really have a true "RMA" process - but we do ask the customer to fill out a form before sending so when an ECU turns up we know who the ecu is from and what we have it for. Also as part of that we give them shipping instructions so they dont get stung with customs/tax.

Set up will be the same as normal - with just the trigger types changed to hall. Use 1JZ VVti trigger mode and VVT mode. Set both trig 1 & 2 to hall, falling edge, pull-ups on. Ill put a pic below. I just googled that kit - man what a load of marketing crap they have on their page. The stock VR sensors are better in almost every way. If you havent already changed the wiring I would suggest you keep the stock sensors. If not then dont be too concerned, not really a down grade - but certainly not an upgrade either... I suspect the real reason they make that kit is because Haltech ecu's dont have a triggerscope so they have trouble diagnosing VR sensors wired incorrect polarity, and their VR conditioner is a bit flakey at low voltages too, again hard to diagnose with no scope. Just because these types of vague uneducated marketing claims annoy me, im going to do a little rant and add some counter argument: The benefits of using Hall Effect Sensors over VR (Magnetic) sensors is a long list: Resistance to Electro Magnetic Interference (EMI) A VR sensor voltage output rises with RPM, this means at higher RPM where noise immunity is more important, is has far higher signal to noise ratio than a hall sensor. Very little or no filtering required Why? if you have the same operating conditions/environment then both hall and VR are going to need the same filtering. In some cases a VR probably needs less due to better SNR. No minimum engine speed to produce a minimum trigger voltage. Important with a slow cranking motor especially with low battery voltage This is the only one that is slightly true, Although all factory VR sensors I have seen put out acceptable voltage from about 50RPM. Not many engines will start when turning that slow regardless of sensor. Triggers precisely on trigger edge and not on trigger voltage threshold position in VR's waveform (which can vary with motor speed) Incorrect, VR's trigger at the zero crossing which is always exactly in the center of the tooth regardless of speed. Hall sensors have what is called a rise and fall time, so they dont trigger "precisely on the edge", they trigger a fixed time delay after the edge - which means the delay angle varies with RPM. Easy Cam and Crank Syncing Huh? The sensors are reading the same pattern and feeding it to the ECU. There is no difference to the signal the micro processor will receive. Precise Ignition Timing There is very little in it either way, but most would say a VR is more accurate due to no rise/fall time delay. No Ignition Timing Float Again same as above. A hall will exhibit more drift with RPM due to rise/fall time - although barely detectable either way. Immune to Target Runout Maybe a valid comment for a DIY triggerwheel that somebody has mounted by eye, but have you ever seem a factory crank trigger wheel runout? Rotational Alignment to target is not critical Not sure what this means? Isnt the kit is meant to fit in the factory holes? High Speed Operation (up to 15khz) How is this an advantage? A VR sensor is unlimited. On top of all that a VR sensor has no electronics inside, so it is more reliable and not sensitive to heat or vibration.

The factory ballast wont work, it will limit the current too much and those injectors get lazy if you limit them too much. You will need to make a heat-sinked bank of 6, 2.2ohm, 50W resistors. Something similar to this one that we make: https://dealers.linkecu.com/BAL6, except ours are only 4.7ohm, 25W so not suitable for these injectors.

A bit hard to tell as there is only one small part of the log where just one lambda is working which is at idle where you have a inj PW of only 0.66ms, I suspect the injector is not very linear down there so is not a great indication of what will happen at higher PW's. Even though you have a VE of 75% in that area , CLL is pulling 20% fuel out to meet the target lambda so that would suggest your VE isnt far off normal in that area.

G4+ Fury has P&H drives so that is fine. G4X Fury doesnt so that would be the same boat as the plug-in.

Yes, the offset will need to be correct for sequential injection to work correctly. Semi-sequential works just like sequential, where the aim is to inject the fuel at close to the most optimum time. It is still only injected once every 720degs like sequential. So, one of those offsets will mean the fuel is being injected 360deg later that ideal. I dont know the 1UZ well enough to know which one is right. To confirm you could wind the engine over by hand and park it at TDC #1 compression stroke. Then pull the cam cover off (or you might even be able to just look down the cam sensor hole) to see where the cam tooth is in relation to the sensor. If the cam tooth is pointing somewhere close to the sensor then the -4 is your correct offset. If the cam tooth is pointing away from the sensor (ie 180 deg out), then your correct offset would be the +356 one.

I wouldnt recommend the 2400 Deka's for a plug-in ECU. The plug-ins only have saturated drivers so you would have to add ballast resistors and the dekas are pretty fussy with current. You can make it work but there are probably easier options.

Try this file, Lambda multiplied by 14.7 r32gtr new CAN.pclr

I had a quick look over the logs. In the log "start.llg" it took 1.4 seconds to sync from when the starter was engaged, it then took a further 2.7 seconds of cranking, before it started (RPM=500). So, a total of 4.1seconds of cranking to start. "start2.llg" was a little better, it took 1.2 seconds to sync, then a further 1.6 seconds to start, so a total of 2.8 seconds. To improve the sync, you could try lowering the 0RPM cell of both trigger 1 & 2 arming threshold tables. Im not sure how much that will help, but it may shorten the time to sync by a few teeth. So I suspect most of the problem is just tune/fuel related. To get a rough idea of what direction you need to head I would suggest making a drastic change to pre-crank prime and first crank enrichment - say double the current numbers. If that doesnt help, then next cold start try halving them. Another thing I notice is your start position signal on DI#3 is not working. I dont know Honda's well enough to know if that is normal for some models or not. But without this working it means your pre-crank prime happens at key-on (when often fuel pressure hasnt built up yet so it doesnt have much effect). So what you could also try before starting is turn the key just to the on position (so fuel pump primes), then turn it off and on again before cranking so you get a pre-crank prime with full pressure. If that helps then you could look at fixing the starter input so that pre-crank prime can happen in start position instead.

No, just Lambda. That one you can just multiply in a custom stream though to make the number read in AFR. - Although it is still going to be labelled Lambda on the screen by the looks.

The MDC is completely standalone and will work fine with a Link Plug-in. The factory diff controller wiring has a TP input for the load reference. The MDC when coupled with a M800 plug-in had an optional RS232 communication which would allow you to use efficiency(MAP/Boost) as load instead of TP only that the factory controller used. You could also log some parameters from the MDC in the Motec ECU over that same RS232 wire. If the MDC has been set up to use ECU efficiency on the lock tables instead of TP, then you will have to change it back to use TP instead. You can log all the same stuff including wheel speeds etc by connecting the MDC CAN to the Link ECU.

70% isnt too far away from where I would would be happy. I often end up with a VE of around 60 for idle. But I wouldnt expect that is so high that you would max out the fuel table. Are you monitoring differential fuel pressure? If you need to reduce your VE numbers then you need to make the ecu think there is less fuel flow or more air flow than there actually is. So you can reduce injector flow rate setting, reduce base fuel press setting, or increase engine capacity. No real advantage on which one you change.

Can you try with the dwell set back to ms mode and put a flat 3ms all over. I think the "cut" might be the dwell getting too short and the MSD doesnt like it.

It needs to be fixed on their end. MAP is sent out 1:1 so cant go negative. MGP is sent out with a 100kpa offset to allow the receiving device to interpret negative values.

The master fuel is just an overall scaling factor so that the main fuel table so it has a usable range. If you have a master that is too big then your fuel table numbers will be very small and you lose some resolution. If your master is too small then you may reach the limit (biggest value the fuel table will accept is 150%) before the engine has enough fuel. Typically a master around 15-20 will do a 500cc injector on a average sized engine with petrol. 7-10ms for a 1000cc injector. 4-7ms for a 1500cc inj. Master fuel trim is just a finer adjustment of the same effect.

Correct, because MAP never goes negative. 50kpa or 7psi is pretty typical idle MAP. But your 7psi at idle won’t still be positive when you subtract 14.5psi from it...

MAP is sent out as a 16bit integer value which means its value can be between 0 & 65535. So if I subtract 100 from the MAP value, anytime the engine is in vacuum (i.e below 100kpa), the ecu cant send out a value less than zero so it will overflow and start counting backwards from 65535kpa. So if your engine idles at 50kpa MAP for instance your boost gauge will show 65485.

For the MAP, what is the channel named in the dash - do they call it "Boost" or MAP? Is it currently displaying a number that matches the MAP reading in PC Link? Im not sure if I can just subtract 100KPa off the MAP value that is sent out as in the data stream they are decoding MAP is not specified to be signed - so it cant read negative values for vacuum.

Easiest to explain in video, so video attached. Flap control.mp4

Attach your ecu config, I will take a look

All data sent out over CAN is in the native "metric" units. This means exhaust mixture is in units of Lambda, temperatures are sent in deg C and pressures in Kpa. Most dash systems have some way to to change the display units on screen. If your dash doesnt then you could set up a custom stream with a multiplier of 14.7 (assuming your fuel is petrol) to receive a AFR value at the other end.

Your onedrive link doesnt work, but I got your email with same files anyhow. It is just a tune issue, the throttle PID isnt quite right and it gets a small oscillation in it after a big change in target which makes the error accumulator increment. Im not really sure what to adjust but I would start by setting the proportional, Integral & derivative to the values in the help file for the Toyota throttle, and do another log to see if it is better or worse. Try these PID values:

It wouldnt hurt to test if you have the opportunity, but I doubt it will be an issue with a relatively freely moving mechanisim like this. The G4X E-throttle controller is rated for 4.4A continuous with shorts bursts to about 10A.

Have a look at the current firmware, most of that is already in the virtual fuel tank functionality (chassis and body menu)