Speeduino build for M20B25

[ADEN]

Steady Idle tip: In the idle area in both VE and Spark maps, Equalize all surrounding cells.See below

[Tini Butscher]

Hi Aden,

Will apply your advise ASAP and will report ;) All sensors calibration are checked.
For the EGO sensor I refered to data shit given by RM Racing Electronics
For GM IAT and flexfuel sensor I choose the rignt brand proposed.
For CLT, I choose BMW E3 325i proposed because it still the stock one

I am wondering about 2 issues I have with my standalone.

#1_Fuel Pump doesn't work. Main relay, Fuel pump relay and fuse (N°11) are OK. On tunerstudio_accessories_Fuel pump is set as seller "board default".
Continuty test have done on loom (between ECU header and relay port).
When I jump Fuel pump on relay port (relay removed), it works.
When I jump pin 3 on ECU header with ground (Fuel pump relay placed on her port), it works.
I guess that the standalone ECU may have a problem and doesn't activate Fuel pump, but don't know why.

May VE table/ spark table/ AFR table do not fit well because of E85 added in fuel tank and also due to 328i injectors.

Required Fuel calculated is around 11.2 with caractristics as follows :
- engine displacement = 2494 CC
- number of cylinders = 6
- injector flow = 250 cc/min
NB : not sure that 328i injector flow is correct!!!


#2_ICV is set as 3 wires on Tunerstudio as follows :
- Open Loop
- Freq (Hz) at 92
- Normal
- 6.1ms

Sometimes it regulates idle well, sometimes it doesn't !!!

PS : I will attached my .msq file later

[ADEN]

Using M50B25NV file as a start...
1- Calibrate TPS, CLT and pressure sensor correctly.
2- Changing ignition settings spark output mode to single channel, trigger edge to Faling and trigger angle to 256.
3- Board layout as recommended by the board seller (most important setting).
4- Control algorithm: MAP.
5- Required fuel: 17.3.
6- Injectors : alternating.
7- Injectors open time:1 ~ 1.2ms.
8- Enabling and Choosing 3 wire ICV.

These basic settings should get you to start it, Then time for tuning idle.

By reading and understanding how it works then fix problems one by one and tuning against target AFR.

If you have a problem please explain it and attach your file along with a log.

[Tini Butscher]

Hi there,
Need help with my 1987 M20B25 (early E30). I turned it to late E30 engine loom, damper, CLT, CID components in order to add a standalone ECU instead of a stock "173 ECU" and above all to delete restrictive stock MAF.

I purchased a RM Racing Electronics PNP ECU which is a "UA4C" speeduino (with VR conditioner and 3 bars MAP sensor included on PCB already).
In the same time, I also purchased with a Wide Band sensor, an E36 VTPS (with support), a GM IAT and GM Flexfuel sensor.

At the moment, I just want to be abble running on atmo and Flex fuel setup. Wish one day to install a turbo set up next.

Engine running very bad. Doubtless in regard to a bad tuning on Tunerstudio!!!

Do you have any advises or ".msq" file using nearly like the best base tune.

Many thanks for your return.

Tini

[ADEN]

I am preparing this project for my m20 special thanks to sk8 for his support and help. But ran into a problem with no available ignition driver ISL9V5036P3. Btw, i am doing a stock single coil setup.

I have access to HGTP14N40F3VL out of the ms41 ecu but way lower amperage rating than the required driver (400V 14A),

Is it going to work like this without problems?

If not, I am thinking of using 2 HGTP14N40F3VL in parallel to run my single coil, So since Q1 and Q4 are sharing the same input from CPU, So by soldering a jumber between :IGN1.1 and IGN1.2 to combine the outputs from Q1 and Q4 and use the stock 55 connector output to coil.

Is it safe to do this?

[varg]

Neat build. My interest is growing for a speeduino for my upcoming 24V build, because of the rather considerable expense ($1300+) for the PNP MS3 unit necessary to do sequential injection, which I am close to considering mandatory due to previous experience with having massive injectors. The really nice thing about the speeduino is that even if there were a reliability issue with it vs MS which I've been running in DDs for more than a decade, it's so cheap you can keep a spare ECU in the glove box

[sk8]

Very good info here, please sk8 tell us more from your experience about reliability and performance of the speeduino comparing to stock ( mpg, idle, power through the power band..)
I'm intersted and on tight budget and this thread brought life to my projects lol.
Subscribed

Reliability seems fine, no issues with the ECU so far other than it sometimes restarts when Im datalogging. I'm guessing that could come from maybe some issues with the computer connection or the Arduino itself. I'm using a cheap knockoff Arduino board.

Can't really compare to stock as far as mpg and power band goes. The standalone ECU won't increase performance any more than a chip tune would on a NA engine. ICV can be controlled by the ECU, so I can set it to idle higher untill it warms up.

I didn't stay NA for long so I never even got a great tune for it when the car was NA. Definitely nice having a standalone ECU with a turbo though, tons of benefits there over stock ECU. I'm logging fuel and oil pressure on top of everything else so that's really nice to have to troubleshoot if things go wrong.

[ADEN]

Very good info here, please sk8 tell us more from your experience about reliability and performance of the speeduino comparing to stock ( mpg, idle, power through the power band..)
I'm intersted and on tight budget and this thread brought life to my projects lol.
Subscribed

[sk8]

The issue I'm having is overboost, I end up hitting my boost cut, maybe this table is looking for boost targets only and not the map targets? So instead of 136kpa I should have 36kpa, need to test that out. Looking at the logs it looks like when I hit 136 the duty cycle stays at 40 (I actually had 40 as the max in my settings at the time of this log). Looks like fuel pressure kind of bounces around. I'm wondering if this is because of my sensor reading weird or the fuel pressure is actually doing that.

Random thing not relating to the ECU. It seems like after turboing the car my oil cap is leaking oil, I have valve cover vented to atmosphere so not sure why it's leaking. Took a bike innertube and made a gasket for it, hopefully that helps.

[sk8]

Got the fuel pressure sensor and boost controller installed and wired this weekend, still waiting for my adapter for the oil pressure sensor. Wiring was pretty straightforward for both.

The boost controller I'm using is a MAC 3 port 35A-AAA-DDBA-1BA. Looks like this and only has two wires. One of the wires went to a switched 12v, and the other one goes to the pin on the ECU that is labeled boost.

I have it plumbed to my wastegate like the image below.

For the fuel pressure sensor I used the holes that are under the arduino on the ecu and soldered wires in for the 5v, ground, and a6. Put a 2.2k resistor between the a6 pin and the signal wire. Not posting pictures because it looks pretty bad so I'll probably redo it sometime.


Trying to used the closed loop and running into issues. Here are my settings.

[sk8]

OEM is the only reliable way to go. The M42's OEM one is well north of $200, so $170 is not too bad overall. It'll last another 3 decades anyway. Hopefully the head gasket was not damaged when the ECU lost sync with crank position. I suspect that my engine's HG was done-in when I had a similar issue (the sensor was OK, but the rubber ring holding the toothed wheel to the hub came apart). It took ~4 months for the HG fully blow after the backfire incident I had, but it never quite ran right during that time, so keep an eye on coolant levels and listen for any "excessive" exhaust noises. You're probably fine though, so hopefully it was just raw fuel in the exhaust getting lit up.

What are your overall impressions of how the car runs on Speeduino, particularly idle and throttle tip-in?

Headgasket is fresh with ARP studs holding it down, so hopefully didn't do too much damage. Idle is great, throttle tip in needs work, but I don't have acceleration enrichment dialed in or know what I'm doing with it. Here is what my acceleration enrichment looks like in settings.


I'll run with this spare CPS I have for as long as it works, and if it fails I'll get the OEM one. Updated the settings in my ignition table to not have negative timing if it was to lose sync again. Here is my ignition table currently, does any of this look too aggressive? I'll be making a set of knock cans soon so I can listen for knock.

[bmwman91]

OEM is the only reliable way to go. The M42's OEM one is well north of $200, so $170 is not too bad overall. It'll last another 3 decades anyway. Hopefully the head gasket was not damaged when the ECU lost sync with crank position. I suspect that my engine's HG was done-in when I had a similar issue (the sensor was OK, but the rubber ring holding the toothed wheel to the hub came apart). It took ~4 months for the HG fully blow after the backfire incident I had, but it never quite ran right during that time, so keep an eye on coolant levels and listen for any "excessive" exhaust noises. You're probably fine though, so hopefully it was just raw fuel in the exhaust getting lit up.

What are your overall impressions of how the car runs on Speeduino, particularly idle and throttle tip-in?

[sk8]

Finally got the car turboed, and started working on getting it tuned. First time I decide to get into boost my crank position sensor decided to fail, which was a pretty scary experience on the road. As I was doing a pull the car felt like it cut out and then backfired super loud. You can see what happened in the logs.

The crank position sensor would short out when I wiggled the wire, and the resistance was ~930 ohms which is out of spec which I think is 530 ohms. I had a spare CPS laying around so went ahead and swapped it, not sure how soon this one will fail, it's definitely not the oem metal one. It read 540 ohm and the car cranked. Does anyone have suggestions for a replacement brand just incase? Seems like the OEM sensor is ~$170 which seems pricy.

Also is it normal to have some sync loss when first cranking the car? Here is an example log from the car cranking.

I plan to wire up oil and fuel pressure sensor, and a boost controller this weekend. Once I get everything working I'll update the thread.

[sk8]

Updated the original post with more details on idle control and trigger angle now that I have a timing light. Got wasted spark working and 60lb injectors running, so going to add details soon. Slowly putting together a turbo kit, so hopefully by summer I'll have boost control working. Car was running fine and I actually took it for a drive. Autotune seemed to dial the AFR in pretty well but with stock injectors, but I was seeing VE table values ~145 at 4400rpm 100kpa which seems really high. I think the reason for that was because I had my injector settings wrong. Upgraded to Simens Deka 60lb injectors so we will see how that goes.

Ran into a really interesting issue when setting up wasted spark. With the stock cap/rotor using the timing light trigger angle was 50, and the car ran fine. When I moved over to wasted spark I had issues cranking the car over, it would just backfire when cranking. Pulled the timing light back out and tried to verify my trigger angle again and it changed from 50 to 276. Not sure what caused this change or why the trigger angle is different from the stock cap/rotor setup, anyone have any ideas? I used the same timing light, and checked both cap/rotor and wasted spark trigger angle on the same day.

[Digitalwave]

One thing I found with two-step on my car is that the stock clutch switch is not well suited for it. If you have the clutch depressed AT ALL, even a little bit, the switch will activate. What I found is that unless I took my foot fully off the clutch pedal between shifts, I was frequently activating two-step when I didn't want it active, just by applying the slightest pressure to the pedal. During autocross, when it's not really practical to take your foot off the pedal (because you are shifting very often), it was a pain in the ass.

For track use and flat foot shifting, that might work fine.

For autocross and two-step, you probably want a bottom mounted switch that will only be active when you really want it to be; clutch fully depressed to the floor.