[Svetoslav]

I think I got that. The strange thing is that it seems to me that the valve I use has a very narrow usable range.
What I mean is that with a fully warmed up engine and the valve opened 100% the idle is around 1500rpm. When the valve starts to close just a little, lets say 80% open, the rpm drop significantly and idles at around 800rpm (which is ok).
However if the valve continues to close the engine dies. There is simply no way the engine to run below that, and in the "limp mode" when it's slightly open it can't .
Which makes me wonder if the valve I have is just too small for my application. It is the Bosch 0 280 140 516.

For the time being I installed the original slide valve and adjust the idle manually to tune in the engine. I'm almost ready to get the car on the road and start tuning it.

[PSIG]

I think I got that. The strange thing is that it seems to me that the valve I use has a very narrow usable range.

A narrow range often means incorrect or an inefficient Hertz, but if it's not tuned around idle, then it's almost irrelevant at this point.

What I mean is that with a fully warmed up engine and the valve opened 100% the idle is around 1500rpm. When the valve starts to close just a little, lets say 80% open, the rpm drop significantly and idles at around 800rpm (which is ok).
However if the valve continues to close the engine dies.

The throttle stop should permit minimum idle rpm, with no IAC flow. Higher rpm if in limp-home or disconnected. I often cap the IAC inlet so initial tuning is not complicated by changes in IAC flow. This of course requires setting reasonable warm idle with the throttle-stop. Many throttle-stop resets may be required during tuning, before the IAC can be used effectively.

There is simply no way the engine to run below that, and in the "limp mode" when it's slightly open it can't .
Which makes me wonder if the valve I have is just too small for my application. It is the Bosch 0 280 140 516.

I have used tiny IAC valves on relatively large engines. I'll bet it isn't set for good idle, which can be difficult without at least initial tuning for idle stability.

For the time being I installed the original slide valve and adjust the idle manually to tune in the engine. I'm almost ready to get the car on the road and start tuning it.

On first start, I warm and tune idle and high idle immediately. This allows both a good setting for IAC, and building the tune foundation from the bottom-up. Many functions (such as active IAC) are not helpful until the tune is ready for it.

I would suggest IAC is mostly for general idle speed range. It does not work well for controlling real-time idle rpm. Idle Advance control or Spark Table timing troughs work much quicker and better for that. In a sense, I only use IAC for general warmup rpm, and warm idle rpm ranges. Other idle stability is with Idle Advance or timing manipulations.

There are several approaches to IAC tuning, but I suggest using the throttle-stop for minimum warm rpm, and IAC only for warmup air during early tuning. Hope that helps.

[Svetoslav]

I think I already tuned the idle at warm engine pretty good by setting it with the bypass valve. It is stable and comes down very nice when I rev the engine.
I could try setting the throttle for the idle, but that's not how it is on the M30 engine and the adjustments are pretty limited. Originally the throttle was always closed and the idle is set by a bypass air valve. There is also a simple auxillary air valve for the warm-up. However this valve is defective, so I decided to change it for an electric valve, like the later M30. They don't seem to have any other way of setting the idle, like the earlier ones with the bypass valve, just the ICV.
But as you said, It wouldn't be а good idea to control the idle only with the ICV. So I think I'll just use the original bypass valve to set the idle and the ICV in open loop to increase the airflow during warm-up.

[Svetoslav]

Started having this weird issue recently where the RPM would jump aound making the engine run poorly.
What could be going on? Looks like noise or something. It doesn't seem to be loosing sync.
Here's a video that I took.

https://youtube.com/watch?v=A0A80OIaEgI ... mBRvwgeoIf

[Svetoslav]

The issue got worse recently. From the logs I can see that it is caused by the crank sensor. Tried different stuff like readjusting it, routing the cable away from the ignition and connecting it to 12v, instead of 5v. The last thing seems to fix most of the issues. There were still a couple of rpm spikes at idle, but it was a lot better and they dissappeared when the engine warmed and I took it out for a ride.
Maybe the sensor is not very good, it was pretty cheap, I will buy a better one to see.

[Svetoslav]

I got my tach working. I Installed an ULN2003 because I need it for fan control and decided to try once more with the tach and what do you know, it works. Although it only shows correct up to 3000 rpm, after that it shows higher than it actually is in tunerstudio. Not sure how to fix it. Tried both Match dwell and the other option, but the results were similar. I'll have to play with the settings a bit more to see.

I will also get a new wideband controller, because the LC-1 seems unreliable to me. Every time I start the car with the same settings it shows similar but different values at idle. Maybe the sensor is not good, but from what I see on the internet there are many people who complain about it.
So I'm thinking of getting the Spartan 3 Lite V2. Does anyone of you use it and do you think it's a good choice?

[PSIG]

I have used 14point7 products successfully. Note the important factor is that the unit responds consistently. Accuracy is not as important. If it reads 1 AFR lean or rich of another unit, that's OK and usable, so long as it does so consistently.

[Svetoslav]

Just a quick question, that bothers me. What will happen if the injector flow is not the correct value? Also is there a way of measuring the flow with the speeduino, like there is with megasquirt?
I'm wondering if I set the correct values for mine, I don't lnow what their flow is, some guy on the bmw forum told me that they're about 210cc/min

[jonbill]

Your RF value will be off in the same proportion, and when you've tuned the VE table, the values will be higher or lower than they would have been with the "correct" RF value :shrug:

[PSIG]

What will happen if the injector flow is not the correct value?

+1 to that. But, nothing 'bad'. If your flow estimate is low, it will be rich from what the calculator expects. And if estimate is high, then lean. So, you allow it to warm fully, then adjust it with VE in that cell. Or if it is generally rich everywhere, adjust Required_fuel (Req_fuel or R_f) to enrich or enlean everything.

Think of it like multipliers. The fuel flow is multiplied by Req_fuel and VE value. You can raise one and lower the other, and other manipulations. This adjustment is actually required in almost every case anyway, as the injector flow can be "correct", but your fuel pressure is a little high, or your voltage is a bit low, etc. Any of several effects can mean the same thing - the estimated fuel flow is incorrect. This is why we tune, and why there are corrections, as nothing is ever prefect or exactly correct.

Having said that, and while every setting has an effect; note that some adjustments and settings become far more important than injector flow, such as dead time (latency) and voltage corrections. Those also affect injector flow, but invisibly, making problems that can be frustrating to find if set poorly.

Also is there a way of measuring the flow with the speeduino, like there is with megasquirt?

Yes, but not required. The setting for Speeduino is 100% duty cycle flow. You can do this by simply powering your fuel system and injecting into a container for specific time and measuring volume. Or injecting to specific volume and measuring time.

So, powering the injector full-on for 15 seconds (either hard-wire or using Hardware Test mode) and measuring 53 CCs in a container would be 4*53 =212cc/min. Or, injecting until it fills to 100cc takes 28.5 seconds, so 100/28.5 = 3.5 cc/sec * 60 seconds = 210cc/min.