Adamw

Experience mostly. There are many factors that come into it, but on a engine with 500cc capacity per cylinder, typical ball park master fuel number would be about 9ms for 1000cc injectors, 17ms for 500cc and 20-25ms for 300ish.

The diagram posted above by Hussain would be valid for any Honda that uses the "OBD1" ecu (CivicLink 95). Except I would not connect the coil ground to Pin 26 as per his diagram. Coils should be grounded to the cylinder head only.

If you have some spare time in the next few days, could you try the G4X back in the Clio and crank it with the spark plugs out to see if you get spark when there is no compression. That would at least back up my thoughts that it is the unstable crank speed that is upsetting it.

Are you on the latest firmware? Your map attached above shows you where on an old version, but the log shows a newer version? Did you upgrade in between? Based on what I see I think this is a firmware problem as the triggerscope is showing the correct number of teeth coming through, but the tooth count is resetting often in the wrong place when the crankshaft quickly slows down (ecu thinks it sees a missing tooth). If you are already on the latest firmware I think this is going to need some adjustments to the code to fix. I will send an email to Link on your behalf and will cc you and ask them to get back to you. Note it is a 4 day Public holiday here in NZ for easter.

Not really anything I can remember that would be hugely relevant to yours (i mostly do ITB race engines). If it where mine I would probably change to alpha N to see if that gives more acceptable control. Sometimes MAP (especially single throttle, big overlap) just isnt a great indicator of air flow. Do fine breakpoints at small throttle openings like my example in this post:

In general it looks like everything is working ok. It doesnt look like the alternator is working. Your ignition table is very retarded in the off boost areas and there is not much VVT action so it is going to feel fairly lethargic like that. As for the boost I would almost be suspicious there is a big boost leak as MAP never goes above 100Kpa.

Is it a G4, G4+ or G4? A bit confusing as your title says G4+, but then has the part number for the G4X, and then you are posting in the G4 section... But in any case the answer is much the same. If the ECU is new then yes it will already have a base map in it. Just do all the pre-start checks etc that are listed in the manual as there may be small settings changes needed for specific car models and things like what type of air temp sensor you have connected. Confirm all basic sensors are reading realistic and calibrate the TPS, MAP and check base timing once it is running. The injector size as well as variables such as fuel pressure are taken care of by the "master fuel" number. It is set at 9ms in the base map which would suggest to me the test car had big injectors. Start with about double that number should get you close to starting up, then just bump it up or down until it runs ok.

Your fuel table set up with 10% breakpoints is not really suitable for ITB's. Because you get a large increase in airflow for the first few percent of throttle opening you need much finer breakpoints at small throttle openings. Once your past about 20% open then your breakpoint increments can get bigger and bigger. Below is an example from one of my ITB engines, do your breakpoints similar to this. Notice at 1000RPM how much difference in fuel there is between 0% and 2% throttle... Also notice at say 3000RPM how little difference there is in fuel between say 40% and 100% throttle. Also, you will need to add a map sensor later if you want to use idle air control, without it the ecu wont be able to compensate for the extra air that the idle valve lets through.

No, leave it as is. The AFR target table is set between 1.00 & 0.95 in all the idle/cruise region so that is having very little influence on the commanded fuel. Backfiring is usually due to lean mixture, not rich. This looks like it would just be due to your fuel table having very small numbers at closed throttle above about 3500RPM.

I wouldnt set the aux up as a fan, as when it turns on it will increase idle speed for the fan load. Instead set the aux up as a GP output then set the condition that you want to turn it on. Look at aux 7 in the GTS base map as an example, it turns on the CE light if temp goes above 102C or if the boost goes above 230Kpa. Red means you are not connected to an ECU. Aux 1 is connected to two pins. Make sure the alternative pin doesnt have a wire connected to it.

If its one of those barrel shaped bosch idle valves then you need to set the minimum clamp to 15%. 100% is fully open, 15% is fully closed, below 15% they start to open again. If that doesnt help then post your tune Ill take a quick look.

Sorry for the confusion Aewon, Vaughan isnt quite right here. Most of the Plug-ins dont have highside drive hardware on board (the only exception is some of the subarus and hondas). The auxes do have a 1.5Kohm pull-up to 12V, so they will drive to 12V when "off", but they wont have enough grunt to drive a relay.

Can you attach your tune. So is the flywheel now 60-2 and no cam sensor?

You realise the "psid" is differential pressure right? It is not very common to run at 100psi differential, I really only see it a little bit in drag racing with methanol & usually mechanical fuel pumps - yes most of the EV14 injectors work ok up to about 125psid. But the typical application that needs 100psi differential is something pretty wild so idle quality is long out the window anyway and deadtimes arent going to matter one bit.

Can you try this. I have lowered the arming thresholds and put in way more crank enrichment and pre-crank prime. 3sgebaserev8.6 Try this.pclr

Not really, I've used the hella one a few times for fuel pumps on race cars with slow/small alternators but only smallish pumps at about 10A. They are rated for 22A, but a few other guys on here that tried them with higher loads said they overheat and reduce output after a short time. In the Megasquirt world the Dorman 902-303 (jeep) or Dorman 902-310 (dodge) are common and are meant to do up to 40A PWM. Never used one personally however.

You can download a PDF of this pic down the bottom so you can print etc. Monsoon LS1 Basic.pdf

That page is for the G4X - although Im not even sure if the X does have an onboard controller. Here is a copy of how the same page read when the G4+ was the current model: http://web.archive.org/web/20180524092028/http://dealers.linkecu.com:80/iPA1_2

The late model TT's have a wideband sensor connected directly to the ECU from the factory, the early model TT's (that the plug in was originally developed on) only had a narrow band sensor from factory. So it sounds like he has a late model car with the factory wideband wired direct to the ECU, but the ECU doesnot have an onboard controller for a wideband. The factory sensor can be re-used, but it will need a 3rd party controller connected between the sensor and ECU.

It may have very minimal effect (accel enrichment may not be as smooth), but at this stage just change it temporarily to see if it effects the run-on. If it helps then we can change the injector timing to a table so that it is only advanced at idle and returns to normal timing all other times.

Can you attach the PDM config too. Your streams 3,4&5 are only sending a few bits each so they can probably be all condensed into a single stream. Can possibly do something smart at the PDM end also to combine output channels into a single byte which would allow stream 1 & 2 to be condensed into 1. Also what is connected to CAN1, what is stream 7, 8 & 9 doing?

That looks like it turns off close enough to instantly to me. There is a small delay in what the laptop displays. I think you will have to experiment with the tune to see if you can eliminate it. The first thing I would try is moving the injector timing later so that end of injection is about 200BTDC, this will mean there is less fuel vapour left in the port. If that didnt help you could also try going richer or leaner around idle to see if that has any influence. Pre-ignition will generally occur easier with very lean mixtures so try going richer as a start.

There is no wideband controller on the Link plug-in ECU, it cannot control a wideband sensor directly like the late model factory ECU's could. You will need to add a separate wideband controller.

I rarely run into any problem and have done some fairly complex CAN networks. Can I take a look at your ecu and PDM config to see if I can offer any suggestions. Are you using both CAN ports on the ECU? You can link CAN 1 & 2 together (have them each end of the bus) to get more flexibility in some cases.

Maybe it was discussing a plug-in ECU? Or an old G4 Xtreme, some of which didnt have E-throttle?