Got my M42 running on standalone (VEMS)

[wazzu70]

Yes, the econo meter output is derived from the wheel speed input to the speedo and engine speed from the tacho. As long as your cluster is working the econo meter will work.

The wheel speed goes to the speedometer first and then goes to the ECU in the factory wiring. Engine speed goes from the crank sensor to the ECU and the ECU outputs a tachometer signal. Just for anyone at home who may be curious how the wiring works :)

I hate the econo gauge because its wasted space. I wish it was easy to convert to oil pressure/temp or something useful like the M3 has!

[hyperE30]

That is simply awesome, would love to have this on my m42, the sound is just so good! keep up the awesome work!

[wazzu70]

Ill try and make a write up or something to make things easier for people.

I've already done all the hard work figuring out the trigger configuration and other things :)

[bmwman91]

I am seriously considering the VEMS system now. From reading around, it looks like MAP / SD mode is generally superior to a MAF setup, except in terms of ease of tuning. MAPs don't really have issues with dirt affecting their transfer function, they respond faster (although I am not sure how important a couple if milliseconds are in this application) and intake reversion is less of an issue. So, while I am not sure how the VEMS system deals with a MAF, it may well be a non-issue. Can you take a screen capture of the tuning software's interface for MAP setup? On top of that, it is a complete product and there's no BSing around with external coil drivers or WBO2 systems. I get the feeling that the LC-1/-2 is not the ideal thing really.

You mentioned a VEMS V4 board...is there talk of such a thing in the piipeline? How do you like the tuning software?

I bought a complete M42 wiring harness a while back with the intent of doing a MEGA-clean stand-alone install a while back. Maybe I can put it to good use in the near future. It looks like the VEMS US site hasn't been updated in 3 years. Are there VEMS forums or anything?

Anyway, this thread kicks ass. So far I think that you are the first M42 guy to even remotely document a stand-alone project. I downloaded VEMSTune and played with it...talk about a learning curve! In contrast to MegaSquirt which is arguably over-documented, VEMS seems to lean the other way lol.

[wazzu70]

VEMS forum: http://www.vemssupport.com/forum/index.php
Unfortunately a lot of the sticky info is from older versions so its not 100% applicable.

As far as I know there is no V4 talk, but Im not exactly in the loop on these things!

I actually hacked my harness and removed all the wires that were unnecessary. It really cleaned things up nicely. That and I know where EVERY wire in the harness goes, which makes troubleshooting easy.

I dont think VEMS can support a MAF. Most people find Speed Density better anyway for the reasons you mention. Its not hard to tune at all. When we can meet up I will just show you how easy it is to tune the VE table with the live analyser.

VEMStune software info: http://www.vems.hu/vt/
Video tutorials: http://www.vems.hu/vt/video/#Videos/...d/Updates_2013

Overall I like the software. Now that things are stable, the focus is on documentation. My biggest gripe is sometimes I am not sure what a parameter controls because its not defined in the help. If you ask on the forum though some of the developers will usually answer/explain. This is the area that is being worked on though!

[wazzu70]

I do like the fact its a complete package, including integrated EGO sensor! Also when you order the unit its tested at the factory and verified working.

One big issue with MS I see is people making mistakes what pinholes they pit the component in, poor soldering which leads to intermittent issues, ect. I didn't want to mess with that!

[bmwman91]

Nice! Yeah, being able to clean up the stock harness seems ideal. It is well constructed, probably better so than an after market one, so being able to leverage it seems like a real boon. What all did you remove from it? The AFM harness seems like the obvious one. I assume that you need to customize the injector harness since it is set up for 2-bank injection rather than full sequential. Did you have to re-run a bunch of sensor grounds to connect at the VEMS unit like the instructions show? For the COP conversion, it seems hard to beat chopping the wires at the ECU end and pulling them through to get the proper lengths, rather than hacking things up at the coil end.

It seems like it would be cool to replace all analog signal wires with shielded wires to add noise rejection. Analog signals in an engine environment have to go through the EMI wringer!

[bmwman91]

I do like the fact its a complete package, including integrated EGO sensor! Also when you order the unit its tested at the factory and verified working.

One big issue with MS I see is people making mistakes what pinholes they pit the component in, poor soldering which leads to intermittent issues, ect. I didn't want to mess with that!

Yeah, the DIY MS modules are no joke. Not super hard, but even for an experienced electrical person it is easy to mix up a component. I can solder 0402 stuff without magnification, but I'd still rather buy a pre-assembled unit despite the DIY v3 MS board being easy through-hole parts. A big reason why my MAF converter hasn't happened yet is because it involves like 80 0402 and 0603 components, and a few TSSOPs, and I have not found a PCB assembly house that can do it for a price that leaves me any margins on the thing without having to drive the cost of it up. There's no way that I am going to hand-assemble those in any sort of volume! I guess I could design one that is 2-3x the size and use TH components and 0805 SMD stuff too...the version with all the tiny parts was like 25x60mm. It was a real bear to design the wiring harnesses to actually fit with the enclosure!

[wazzu70]

Yes I removed the AFM harness. There were other things I removed too that were more minor components. I also made a new injector harness like you say so I could run sequential injection.

I used all the sensor grounds and bundled them, and I bundled all the ECU grounds too. I can take a more specific picture. In the pictures above you can see the bundle of ground wires before I crimped them (brown wires).

I could have easily soldered the MS board, but just because you can does not make it a good idea! I would have used VEMS even if I had to solder the components on just because its an integrated design without a ton of add on boards. However, I was really glad I could focus my time on things like wiring harness!

[decay]

I am seriously considering the VEMS system now.

does this mean work has ceased on your Motronic MAF? because i want that :(

[bmwman91]

does this mean work has ceased on your Motronic MAF? because i want that :(

Not sure. I need to find a contract manufacturer to make the things. Building them myself...well, my day job pays me considerably more per hour lol.

For my own purposes, I want to go stand-alone since the Motronic is a functional but aging platform and it forces you to pay a tuning premium to the few that have the knowledge and tools to manipulate it.

[decay]

Not sure. I need to find a contract manufacturer to make the things. Building them myself...well, my day job pays me considerably more per hour lol.

yeah, i imagine, same boat here.

For my own purposes, I want to go stand-alone since the Motronic is a functional but aging platform and it forces you to pay a tuning premium to the few that have the knowledge and tools to manipulate it.

well, i've already laid out for a MarkD chip, so i hope you do farm these out at some point- really want to see how the engine breathes without the AFM in front of the throttle bodies.

[bmwman91]

yeah, i imagine, same boat here.



well, i've already laid out for a MarkD chip, so i hope you do farm these out at some point- really want to see how the engine breathes without the AFM in front of the throttle bodies.

Part of me wants to develop a DIY kit that uses big surface mount stuff and through-hole parts. The existing design involves ~80 0402 (1mm x 0.5mm) R & C parts, and some fine pitch TSSOP stuff that I have to use a binocular microscope to solder. If I went to bigger stuff, the board would probably occupy 4x the space. Maybe I am just anal about making this thing super small and nobody would care though.

Truthfully, the M42 stands to gain basically nothing from a MAF conversion. Even my 205bhp MM 2.1L only really stands to gain ~5bhp from a MAF and M30 air box (per MM). As counterintuitive as it may seem, the AFM door poses basically no restriction. The door takes like 0.5lb to push open, and it's surface area is ~4inn^2, which means roughly a 0.125PSI drop across it. The throttle body and inlet ports pose much more resistance than that. The M20 guys get boned because their AFM has the same amount of crossectional area as ours, with a considerably larger engine. The gains that they get from a MAF are about the same as what they get from an M30 AFM conversion simply because they are choked up by their stock AFM's small flow passage. Now, you DO get better throttle response with the MAF, but with the Motronic you are still not getting the best response that you can. The thing was designed for a slow AFM and a stand-alone ECU will be what really makes the difference. And then in that case, a MAP-based setup would be better in my opinion. MAFs are nice because they save you some tuning time, but they are sort of a pain in the ass because their transfer function changes as the thing gets dirty and whatnot. MAP is superior in almost every way as far as I can tell. Which actually relates to an idea that I have had for years about converting the AFM to a MAP setup. That's a product that I would feel better about selling truthfully.

decay, are you ever in the south bay? I want to see your ITBs! My wife and I do stay up in Point Richmond from time to time as well (her parents live there). We could grab beers and make fun of the hippies in Berkeley or something.

Anyway, sorry about the hijack wazzu. Your VEMS adventure has me all excited about the possibility of a stand-alone install now.

[wazzu70]

Its not a hijack, its all good info and things people considering a standalone want to know.

In the EVO world there are MAF to MAP conversions, so you may want to research there and see what people think about the change.

This is the first car I have driven with MAP based load calculation (speed density) but it is really responsive and smooth. No complaints here!

I still want to post up pictures but I was busy this weekend. I will get some engine bay shots and pictures of how I did the grounding.

[decay]

Part of me wants to develop a DIY kit that uses big surface mount stuff and through-hole parts. The existing design involves ~80 0402 (1mm x 0.5mm) R & C parts, and some fine pitch TSSOP stuff that I have to use a binocular microscope to solder. If I went to bigger stuff, the board would probably occupy 4x the space. Maybe I am just anal about making this thing super small and nobody would care though.

look at the big-ass AFM we're replacing... i wouldn't care about size, i can find somewhere to hide it.

Truthfully, the M42 stands to gain basically nothing from a MAF conversion. Even my 205bhp MM 2.1L only really stands to gain ~5bhp from a MAF and M30 air box (per MM). As counterintuitive as it may seem, the AFM door poses basically no restriction. The door takes like 0.5lb to push open, and it's surface area is ~4inn^2, which means roughly a 0.125PSI drop across it. The throttle body and inlet ports pose much more resistance than that. The M20 guys get boned because their AFM has the same amount of crossectional area as ours, with a considerably larger engine. The gains that they get from a MAF are about the same as what they get from an M30 AFM conversion simply because they are choked up by their stock AFM's small flow passage. Now, you DO get better throttle response with the MAF, but with the Motronic you are still not getting the best response that you can. The thing was designed for a slow AFM and a stand-alone ECU will be what really makes the difference. And then in that case, a MAP-based setup would be better in my opinion. MAFs are nice because they save you some tuning time, but they are sort of a pain in the ass because their transfer function changes as the thing gets dirty and whatnot. MAP is superior in almost every way as far as I can tell. Which actually relates to an idea that I have had for years about converting the AFM to a MAP setup. That's a product that I would feel better about selling truthfully.

i get what you're saying here, but where the MAF would help me is tuning a turbo. you cannot trust an AFM for that *at all*, but a MAF is a much better source of truth for airflow, and i'm relatively confident i could use it to run a low-pressure (sub-10psi) setup, which is all i want.

decay, are you ever in the south bay? I want to see your ITBs! My wife and I do stay up in Point Richmond from time to time as well (her parents live there). We could grab beers and make fun of the hippies in Berkeley or something.

i don't go to the south bay much anymore- i work remotely for a company in Atlanta now, out of my house, so i don't really have much reason to venture out of E'ville. i know good beer gardens around here if you're in the area. but life is constantly in flux; i might be moving to the southern peninsula early next year, or possibly sooner.