Constantly having to tune

[dlenardu]

I just converted a 35hp twin cylinder motor and it seems like I am constantly having to tune it. I get it running really good one day but then a couple of days later it runs like crap. I make some more adjustments and get it running great again then later on in the day it won't run right. It also doesn't seem to have the power it did before. The machine ran great on the carb before I did this. It had loads of power but it drank tons of fuel. I did this swap hoping to mainly gain more fuel economy and hopefully a little more power. I do not have an o2 sensor so I'm not sure what kind of AFR I am at, I have been doing "plug chops" and reading the plugs. What am I doing wrong?

[R100RT]

I have empathy for that situation, have experienced similar challenges. However, my recognition (and orientation) is that the tuning process is totally scientific, and for each and every behaviour aspect (and problem) there is a reason or cause.
I'm not 100% successful in my tuning efforts on my 2 lung bike engine, however I've witnessed good progress and feel I understand now what my issues were at the end of last seasons sessions. Hope that helps some.
Most likely you will be requested to post a data log of the engine running, in order to see what is going awry, that would be my advice to gain more advice from the forums.
Cheers,
Lorne.

[grippo]

If you are losing tune from day to day, the most likely causes are:

1. The inputs you tune in are not being burned into flash, so they are gone as soon as you shut off the engine

2. There is something mechanically wrong going on - like a flaky fuel pump, or flaky injector, etc, and that is what is varying from day to day

3. If the problem is related to the outside temperature, make sure you tune your VE and spark tables only when the engine is thoroughly warmed up, then tune for cold starting without ever changing VE and spark tables.

Do not look for a magic tune that is going to fix this. If the car runs fine one day when it is thoroughly war, then it should run the same the next day when warm. If the air temp is really different on the two days, it could have an effect, but not a night and day difference.

[800vtwin]

If you do a plug chop - which of the 144 cells are you checking??? (widebands are cheap)

What was the air temperature on those days? Whats the air temp corrections set at? Did you add some old fuel or change fuel?

[dlenardu]

If you are losing tune from day to day, the most likely causes are:

1. The inputs you tune in are not being burned into flash, so they are gone as soon as you shut off the engine

2. There is something mechanically wrong going on - like a flaky fuel pump, or flaky injector, etc, and that is what is varying from day to day

3. If the problem is related to the outside temperature, make sure you tune your VE and spark tables only when the engine is thoroughly warmed up, then tune for cold starting without ever changing VE and spark tables.

Do not look for a magic tune that is going to fix this. If the car runs fine one day when it is thoroughly war, then it should run the same the next day when warm. If the air temp is really different on the two days, it could have an effect, but not a night and day difference.

As far as I know everything is being burned in cause when I connect up to microsquirt all of my VE table settings are still the same. I hope it isn't something mechanical, all of the components are brand new. I have a guage on the pressure regulator and the pressure stays where it should. I have only been setting the VE table when the engine is completely warm. I'm running fuel only.

If you do a plug chop - which of the 144 cells are you checking??? (widebands are cheap)

What was the air temperature on those days? Whats the air temp corrections set at? Did you add some old fuel or change fuel?

This machine is governed to around 3600 +/- 50 rpms with a load on. I set the rpms in my VE table to 3800 rpms and I watch the cells at around 95-100 kpa and in around the 3600-3700 rpm with the blades turning (under load) I would have to install 2 widebands on this machine because it is a twin cylinder and the way the exhaust runs I could not measure both cylinders with one. Of course I guess I could always just measure one.

As far as air temp goes, yes it was different air temps. Where can I set the air temp corrections and how am I suppose to do it? I thought microsquirt already knew what the corrections should be based on GM sensors.

[grippo]

I believe air temperature corrections are under Sensor Calibration. There is a built in air temperature correction and this should work, provided the sensor is properly placed and accurate. But I would not change anything right away. What I would do first is to try to duplicate the problem but take short datalogs. Pick a day, let the engine warm up, then datalog for a minute or so. Then wait for another day that is colder/ warmer and repeat. Basically wait for a day when it will run terribly. Then compare the 2 datalogs. If the pulsewidths are the same, then likely something is wrong with the injector electronics or mechanics, or the fuel pump. If the pulsewidths are not the same then look at the data that goes into the pulsewidth: ve, gammae, barocor, map, etc. Something will be different. It could even be a bad tps signal that is falsely looking like the engine is accelerating.

[800vtwin]

No- the GM sensor you speak of is the pre set calibration. But redo calibration for what ever sensor you run under tools/thermistor tables. This only sets how the ecm will decifer the sensor. (what resistance = what temperature) and nothing to do with corrections.

The air correction is under IAT-Based fuel Correction under Other tuning. But like Al said- test/find out whats wrong first. Maybe set these all to 0 for now.

I have seen the accel enrich cycling from a bad tps learn (done with battery charger on or low battery) or from noise/ too low of tpsdot setting. (under accel wizard) You can see if the accel enrich is cycling from the main page indicators.

You can use 1 wideband and get 2 bungs. Put one in each pipe. Start with single table tuning. Tune each cylinder and export/ save the fuel table. Then switch to dual tuning and import the correct table to each cylinder.

Make sure the WUE is turning off and set right. Default setting are very very rich for small engines/ vtwins.

Also - make sure your injector timing is set up correctly. This has a huge difference on idle quality and low rpm running. I haven't seen your msq but you may need to delay a cycle which would be a setting of atleast 50%. (injection fires on tach pulse)

[dlenardu]

I have attached the .msq I am not running a tps. This machine will live around 3700 rpms or so.

grippo: I will try to get some datalogs. Tuesday we will be doing a test so I will get a datalog and then try to get another datalog when the temps are different.

800vtwin: I checked the IAT settings and they were all set to zero.

[800vtwin]

Does it start ok?

If not - change cranking trigger to calculated. (under Base Ignition Settings) and set next pulse tolerence/ cranking higher.(90-200). And skip pulses to 1.

You must have tiny tiny injectors or really low fuel pressure. (the req fuel value is huge)

from what I see - the microsquirt has no way of knowing what stroke you are on. Since you have no input to dictate which tach is TDC compression, and you have the injectors alternating.

One time you start it- it could set up to fire injectors ATDC on the intake stroke and then the next time you start it you could be firing injectors ATDC on the compression stroke. The wrong injector would be firing for each cylinder and completely changing your required fuel table (injectors fire at tach tooth- you have tach/delay teeth set 9 teeth (90*) past gap. But no way of saying where in the 720* cycle. (90* or 450*) And since #2 injector fires 360* from #1, if you have an odd angle engine- when started with the #2 injector giving fuel to #1 - it could be way off in the timing.

Just Switch to simutanious injectors - and retune ve table. Does this engine have a common intake? (it will still work either way) And then play with the injection delay to around 50-75% setting.

[dlenardu]

It starts good, especially when warmed up, but sometimes it has to turn over for a while. I would like it to start better.

I thought that the required fuel value was a bit large. This motor was suppose to be making 35hp from the factory with the carb. I figured that with the fuel injection I would pick up a little horsepower so I estimated it at 40hp. According to the megamanual I needed 12 lb injectors. I have 14.4 lb injectors and I am running them at 30 psi to give me the 12lbs. Do I need to change this?

I wondered myself how microsquirt knew when to inject the fuel. I have a 36-1 tooth trigger wheel and I set it up according to DIY's instructions. I set it to TDC on the number 1 cylinder and had the gap 9 teeth ahead of the sensor.

Yes this unit has a common intake. I gutted the carb and used it as the throttle body. I machined the original manifold for the injectors and I am running port injection.

If I change to simultaneous injection will that prevent it from injecting out of time? Will I be wasting alot of fuel since both will be injecting at the same time? One of the reasons we switched to fuel injection was to try and get some kind of fuel economy but not loose any power. With the carb the unit burned 3 gallons per hour. We are hoping to get down to 2 gallons an hour and hopefully a little more power as well.

I will read the manual some more on the injection delay, I'm not sure I fully understand it.