[Success Story] 1998 BMW V12 engine M73 #55

[Lambo97]

OK that is better. I thought the ETBs were going to be a problem. Has everyone settled on the Nissan's for the V12 or is there any other close contenders for reliability and air flow?

Also Andrey my question about the sequential vs Batch fire was related to avoiding 2 microrusefi's to run the motor, and also performance. Obviously as the motor RPM gets faster the injector firing pulse gets to a point were it's open most of the time so sequential injectors is not a big concern. Also my guess is that running 2 banks of 3 coils thus 6 total in a wasted spark setup would be a cost savings over individual coil packs.

My question is do you think the proteus board is the right way to go with the v12 or is there a more practical rusEFI board to do the job?
If you think the Proteus board is a good way to go, when do you think there will be another batch released?

[mck1117]

OK that is better. I thought the ETBs were going to be a problem. Has everyone settled on the Nissan's for the V12 or is there any other close contenders for reliability and air flow?

The Nissan throttles are of similar size as the stock BMW throttles, and nearly bolt up, so they're a reasonable option at least.

Also Andrey my question about the sequential vs Batch fire was related to avoiding 2 microrusefi's to run the motor, and also performance. Obviously as the motor RPM gets faster the injector firing pulse gets to a point were it's open most of the time so sequential injectors is not a big concern. Also my guess is that running 2 banks of 3 coils thus 6 total in a wasted spark setup would be a cost savings over individual coil packs.

My question is do you think the proteus board is the right way to go with the v12 or is there a more practical rusEFI board to do the job?
If you think the Proteus board is a good way to go, when do you think there will be another batch released?

I don't think anybody seriously recommends trying to run 2 MREs on one engine - that was mostly a publicity stunt

The best way to go is to run a Proteus. It easily supports the v12 (with whatever fuel/ign mode you want) and dual throttles all by itself.

[AndreyB]

Myself and Matt are two different humans. We do not live together (yet). We are many miles apart.

Andrey has a v12 and Matt has a (shrinking) number of Proteus boards in stock.

Proteus is the best rusEFI ECU for any v12.

No one is planning any additional research into better ETBs, we are happy with Nissan from 3.5 v6. Do not confuse with similarly looking Nissan from 2.5 4 cylinder.

[mck1117]

We do not live together (yet).

Time to buy land and build a track on it?

No one is planning any additional research into better ETBs, we are happy with Nissan from 3.5 v6. Do not confuse with similarly looking Nissan from 2.5 4 cylinder.

That said - pretty much ETB that's mechanically compatible (interpret as you will) with the engine will work fine. I don't think we've yet found an ETB we couldn't control.

[Lambo97]

Wow I didn't realize Matt was MCK1117. So Matt I saw someone with a build video on YouTube is that you?

Also what is the current rate for 2 boards as we are looking to follow along with Andrey (a few steps back). Will there be any additional surface mount parts required or is the board mostly complete just requiring external cabling etc...

Thanks for your help,

[mck1117]

So Matt I saw someone with a build video on YouTube is that you?

Was it me? ¯\_(ツ)_/¯

This is my YouTube: https://www.youtube.com/channel/UCRgyNICLVtFrdki4bOugULA

Also what is the current rate for 2 boards as we are looking to follow along with Andrey (a few steps back). Will there be any additional surface mount parts required or is the board mostly complete just requiring external cabling etc...

Where are you in the world? I have a few Proteus 0.4 boards left in the batch up for grabs. Pricing is here: https://rusefi.com/forum/viewtopic.php?p=37503#p37503

The fully spec'd option (with connectors) is fully assembled, ready to plug in to a car (and tested on my car if you pick this option!), and includes plugs and crimp pins to build up a harness.

[Lambo97]

Hahaha I went through the thread end to end I see Audrey was the one soldering the boards. We are In Canada, I'll start a new thread with our cars also to watch the progress. The engines will be here in 2 weeks after which we will strip them down mechanically check everything and then rebuild. They are M73N from 1998.

The M73 comes with 2 of everything, maybe Andrey can answer this one for the V12 as I'm not sure if when using the Proteus board with the M73 are you using one VR cranks sensor in conjunction with a Cam sensor? If using VR are you using both? That would determine if we need to purchase the $15 VR sensor input addon and how many.

So Matt am I to assume you currently have 2 v.4 boards available with connectors?

[AndreyB]

@mck1117 please do not respond on Proteus stock on this "1998 BMW V12 engine M73 #55" thread if that's OK with you?

@Lambo97 my devils are not happy with too much off-topic discussion on this "1998 BMW V12 engine M73 #55" thread. I am only a human with too many weaknesses and preference for nicely structured information.

[mck1117]

Hahaha I went through the thread end to end I see Audrey was the one soldering the boards. We are In Canada, I'll start a new thread with our cars also to watch the progress. The engines will be here in 2 weeks after which we will strip them down mechanically check everything and then rebuild. They are M73N from 1998.

So Matt am I to assume you currently have 2 v.4 boards available with connectors?

I've done the design/assembly/distribution on Proteus so far. Send me a PM and we'll get you squared away!

The M73 comes with 2 of everything, maybe Andrey can answer this one for the V12 as I'm not sure if when using the Proteus board with the M73 are you using one VR cranks sensor in conjunction with a Cam sensor? If using VR are you using both? That would determine if we need to purchase the $15 VR sensor input addon and how many.

99.8% of engines will run totally fine without a cam sensor - running wasted spark and batch fuel (read: firing every injector every 360 degrees, instead of every 720 degrees). That means you can run with just the crank sensor (one of the two VR), and ignore the other. BMW fitted two sensors offset by 60 degrees so that they could use a pair of 6 cylinder ECUs, each running its own "6 cylinder" that happens to be conjoined with another at the crankshaft. You're probably 100% fine with just a single VR input, using either of the two stock crank sensors.

[AndreyB]

The M73 comes with 2 of everything

M73 comes with 2 of somethings and NOT two of many other things. I believe there is only one engine coolant temperature sensor, only one intake temperature and only one cam sensor? I could be wrong.

https://github.com/rusefi/rusefi/wiki/HOWTO-M73-v12-on-Proteus has many useful details

There are definitely two VR crank position sensors but Proteus board would only use one of these two VR sensors, the other one would be completely unused.

[Lambo97]

Thanks Andrey for this info. Just for an updates sake, have you started a tune yet? Would love to see another updated video of how it's running today. Obviously everyone has a life so no pressure.

Thanks, again

[AndreyB]

No clear plans for this v12 so not spending any time on it.

Next time I would touch the v12 could be once/if we have transmission control - but again, this v12 would not be first transmission control test mule for me.

[AndreyB]

Just listed by used ECU+harness on eBay https://www.ebay.com/itm/333746413420

It works if you have five black and five grey ECU plugs like pictured, I believe that earlier engines had a bit different wiring.

Pictured with 134 pin breakout boards https://rusefi.com/forum/viewtopic.php?f=4&t=1524

See also https://rusefi.com/shop/

[mk e]

No clear plans for this v12 so not spending any time on it.

Well get that engine a home man!
https://philadelphia.craigslist.org/pts/d/philadelphia-2000-bmw-z3-parts/7175531912.html

[AndreyB]

I am pretty sure that starter power stud is not supposed to be spinning with the nut Grinder time?

[kb1gtt]

Grinder seems like a good plan

ItalianToastedSubGrinderPrint.jpg (447.39 KiB) Viewed 16733 times

[AndreyB]

BMW has fancy electronic start button and starter has some sort of output as third wire.

rusEFI so far was just cranking for 7 seconds ignoring starter output. Could that have caused starter death? Do we have other theories for what could have contributed to starter departure?

[mk e]

BMW has fancy electronic start button and starter has some sort of output as third wire.

rusEFI so far was just cranking for 7 seconds ignoring starter output. Could that have caused starter death? Do we have other theories for what could have contributed to starter departure?

7 seconds seems like a very long time, so yes that could be an issue. Do you have a state=running or rpm>XXX = running or anything like that so you could 7 second OR running? Normally you'd also want (neutral OR park) AND brake pedal also....I know its early is the work but best to plan for what you'll need.

[AndreyB]

Gotcha, will implement "do not crank" if RPM > cranking threshold.

[mk e]

Gotcha, will implement "do not crank" if RPM > cranking threshold.

Perfect for the test bench! Maybe set the cranking threshold high enough that it will still crank if its coughing a little and trying to start?

Don't let it out into the wild without a brake and trans/clutch interlock or pretty much for sure someone will drive through a garage door or wall

[AndreyB]

Turned out that ""do not crank" if RPM > cranking" is already there but I've destroyed a second starter I believe that's a bug in start/stop where starter is engaged forever. Waiting for debugging interface for Proteus since the defect might be on the edge between HW and SW.

In more optimistic news second M73 Proteus harness is ready and tested.

[AndreyB]

I can now cranking reliably but there is a new unexpected behavior:

if I just tap "Start" button I often get invalid 20000 RPM reading which stops cranking right away. If I hold "start" button I get more proper cranking. That's unexpected since debounce code is supposed to make tapping and holding act exactly the same.

Just for reference, at start of cranking batter falls down to about 8v for a moment but then goes to 10v for duration of cranking.
I am thinking to implement something VBatt averaging and throw a fatal error if VBatt falls below specified expected value for too long.


https://rusefi.com/online/view.php?log=68
https://rusefi.com/online/view.php?log=69

image.png (95.96 KiB) Viewed 16465 times

[mck1117]

Just for reference, at start of cranking batter falls down to about 8v for a moment but then goes to 10v for duration of cranking.
I am thinking to implement something VBatt averaging and throw a fatal error if VBatt falls below specified expected value for too long.

what's the threshold for "low battery"? When looking at battery voltage with a scope, I've seen <5 volts for a few MS upon initial engage of the starter, then up to 9-10 volts while cranking.

[AndreyB]

what's the threshold for "low battery"? When looking at battery voltage with a scope, I've seen <5 volts for a few MS upon initial engage of the starter, then up to 9-10 volts while cranking.

would not be looking at instant value, would be "CRITICAL you battery was below 7 volts for 500 ms"

[mk e]

what's the threshold for "low battery"? When looking at battery voltage with a scope, I've seen <5 volts for a few MS upon initial engage of the starter, then up to 9-10 volts while cranking.

would not be looking at instant value, would be "CRITICAL you battery was below 7 volts for 500 ms"

So the long running "standard", measuring with a meter not a scope, is 8V. When they started switching gauges over from mechanical to electronic they were all 8V....my '84 ferrari gauges are 8V. Most of the pre-done dwell tables and such I've seen start at 8V. I know the motec and haltech ECUs had an 8V output ....that is now 6.3V on the motec, not sure about others. Cranking under 10V usually means battery time, measured with a meter but I have seen down around 8 in winter on a car that just barely comes to life, maybe.

The ms question though I don't know other than most ecus are still 5V so I assume bad things would happen below that for even ms times....but the GM injector correction tables I downloaded from injector dynamics and use in my ecu are 4-20V...its a 17x33 table so they invested a lot of time and memory into it.....but round number wise they need the 4V row to have correct times at 4.8V where I thing the chip shuts down.

In theory an electric motor makes infinite torque at 0rpm while pulling infinite current...wires sizes limit the current though. The size/length of the starter wire must set the min battery V? Which or course makes me realize that by not really thinking about this I probably screwed up the ferrari wiring as the battery is up front , engine rea/mid, ECU truck so I very cleverly ran a huge wire to the starter then pull the ECU power are the starter lug rather than running a separate wire to the battery My point was you should probably only see low voltage on the starter line, not the ECU line if everything is done correctly.

A lot of racing batteries are now 16V btw which changes things again....

Edit: the 8v gauges I've seen are electric, not electronic as in analog not digital....temp and pressure gauges. Every tach or speedo I've seen had a chip of some kind in it but I never looked clos enough to know if the mechanism itself was 5 or 8V

[AndreyB]

Facts:
I have a paddle of dirt mixed with gasoline under the engine.
Many / all injectors on both sides are wet
I have fuel pressure regulator with vacuum line disconnected since I've moved to Nissan ETBs

Theory: lack of vacuum line causes pressure regulator not to regulate, excessive fuel pressure is sipping via old o-rings?

Action plan:
Find a source of vacuum on the engine?
Find an inline mechanical fuel pressure gauge?

[mk e]

That's kind of odd.

What kind of pressure regulator? OEM? Aftermarket? Mechanical?

When you say the injectors are wet....as in you are sure they're dipping and making the puddle? its not unusual to see signs or weeping but actually dripping would mean 1 or more failed O-rings or corrosion in the o-ring seat area, something like that and it needs to come apart as regulator working right or not there is no way the pump should be able to blow past the o-rings. A fitting somewhere is more likely thought which is why I asked if you're sure they are dripping.

[AndreyB]

Ethan was stopping by for unrelated reasons so now I have his professional master mechanic confirmation that yes o-rights are leaking at 30PSI. he says either really contaminated fuel or miracle of simultaneous failure due to age.

[AndreyB]

Jared is wondering what is the history of Ethanol in US gas? In NJ all gas is 10% ethanol is it? Could that be relevant?

[mk e]

Jared is wondering what is the history of Ethanol in US gas? In NJ all gas is 10% ethanol is it? Could that be relevant?

It shouldn't both the o-rings....it can corrode alum as it holds water. Its been around a long time in some places but nearly everywhere starting like...2006? when they linked the MTBE they were using to all sorts of bad stuff. Going back to like 1995 I filled up....I forget which state, on my way back from racing at dayton bike week and bam! fuel mileage was down 20%? iirc, winter gas blends are different in different places and this one has a ton of alcohol I assume. Point is the OEMs have been designing for it for quite some time. Its a bit of a problem for people like me with aluminum fuel tanks in a vehicle that sits long periods of time but generally considered safe for older vehicles.