SpecE30 and Megasquirt

#41

Scott, I really DO understand how difficult and frustrating that can be. My point is that if you start with used components on a race car build, it’s false economy. Those unknowns are going to cost you track time. Expensive track time.

If you chase problems by swapping other used parts, that compounds the difficulty.

Can we at least agree on that?

Again, this rule change wouldn’t have an adverse affect on me. My concern is for the health and stability of the class.

What happens when the test bed comes up with a map that is just right on those few cars, but when it is deployed class-wide we have a few engines pop because they run too lean on it? Would they have popped on a stock DME? Maybe. Will they blame the Megasquirt? Definitely. Is that productive? Nope.

#42

I agree that chasing problems with used parts is risky. But a new harness, DME, alternator and AFM would be very expensive. The culturally correct approach is to test the various parts to see if they are within spec, clean/flow the injectors and head for the track. And that works, until it doesn’t.

Then one starts understanding that the specs are loose. The inner range of the specs are “good” and the outer range of the specs “suck”. Or one finds out that what seems to test fine in the garage becomes intermittant when on the track.

All the sudden replacing everything DME and forward with new parts starts gaining appeal. So yes, I agree with what your saying. New parts are a big step forward, but we’re kind of talking apples and oranges. You’re talking about reducing the liklihood of problems in the first place, and I’m talking about having better diagnostic info. It would be a bummer to spend >$3k on the new parts listed above. I’d much rather spend $500 and have the tools to identify problematic used parts should they start to suck. I’m not suggesting that we choose one course of action over another, or even compare the relative merits of one over another. I’m all for new parts, but at this rate I’m never going to get my credit card bills back under control.

I agree that it’s possible that there will be problems and folks will blame MS. Nothing is impossible.

Keep in mind tho that I’m not saying “hey we should require MS on all of our cars”. All I’m saying is “hey this is interesting. There could be some real benefits here. Maybe we should test this idea”.

#43

I’ll jump in here (at my peril) with a few comments:

I can attest to having a 2 year battle with the Motronic 1.0 on my 85 325e. I have the book mentioned above, along with probably every Bosch FI book even remotely related to early Motronic published by Bentley as of 2005. I can’t say that any of them is terribly helpful in doing more than explaining the basic principles of operation. I was unable to find any schematics for the ECU, which made matters worse. Ultimately the AFM was bad and I missed it as a result of my own stupid mistake, but it cost me a bundle in my time & aggravation. Bending the wiper on my original unit and buying a ‘refurbished’ AFM were both totally unsuccessful but led me to ignore some fairly obvious information (in retrospect). During the process, I learned that many BMW trained technicians, even ones I know personally to be competent mechanics, are unable or unwilling to spend the necessary time and energy to fix troublesome problems. I built & used a megasquirt II system & wideband AFM to help troubleshoot the problem. Yes, a dyno would probably have been just as good & gone a long way to solving the issue at much lower cost, but you can’t always duplicate the failure.

That being said, perhaps a high speed datalogger would accomplish much the same purpose. I believe that Tyco Electronics has not yet obsoleted the 25 & 35 pin motronic connectors, so they can be bought direct from Tyco (in bulk) with none of the typical markup. I researched this awhile back because I was seriously considering replacing my Motronic 1.0 DME with MS-II. We could use megasquirt or another data logger for measuring these signals without replacing the stock DME. If a male/female connector arrangement is built up to be inserted inline with the harness, it could be a handy tool which could tell us what we need to know without the worry about changing ECUs & the associated issues. Another bonus that I always like - no need to chop the harness!

Another clarification, if I may. Megasquirt has a MAP sensor built onto the board, so there’s no need to pick one. If I am not mistaken, the issue of modifying the intake manifold for the map sensor is also moot because we can put a T in the vacuum line for the fuel pressure regulator. I know this as a result of my data logging efforts on the M20B27.

#44

[quote=“Herdrock” post=57873]I’ll jump in here (at my peril) with a few comments:

I can attest to having a 2 year battle with the Motronic 1.0 on my 85 325e. I have the book mentioned above, along with probably every Bosch FI book even remotely related to early Motronic published by Bentley as of 2005. I can’t say that any of them is terribly helpful in doing more than explaining the basic principles of operation. I was unable to find any schematics for the ECU, which made matters worse. Ultimately the AFM was bad and I missed it as a result of my own stupid mistake, but it cost me a bundle in my time & aggravation. Bending the wiper on my original unit and buying a ‘refurbished’ AFM were both totally unsuccessful but led me to ignore some fairly obvious information (in retrospect). During the process, I learned that many BMW trained technicians, even ones I know personally to be competent mechanics, are unable or unwilling to spend the necessary time and energy to fix troublesome problems. I built & used a megasquirt II system & wideband AFM to help troubleshoot the problem. Yes, a dyno would probably have been just as good & gone a long way to solving the issue at much lower cost, but you can’t always duplicate the failure.

That being said, perhaps a high speed datalogger would accomplish much the same purpose. I believe that Tyco Electronics has not yet obsoleted the 25 & 35 pin motronic connectors, so they can be bought direct from Tyco (in bulk) with none of the typical markup. I researched this awhile back because I was seriously considering replacing my Motronic 1.0 DME with MS-II. We could use megasquirt or another data logger for measuring these signals without replacing the stock DME. If a male/female connector arrangement is built up to be inserted inline with the harness, it could be a handy tool which could tell us what we need to know without the worry about changing ECUs & the associated issues. Another bonus that I always like - no need to chop the harness!

Another clarification, if I may. Megasquirt has a MAP sensor built onto the board, so there’s no need to pick one. If I am not mistaken, the issue of modifying the intake manifold for the map sensor is also moot because we can put a T in the vacuum line for the fuel pressure regulator. I know this as a result of my data logging efforts on the M20B27.[/quote]
Interesting stuff.

Would you be willing to look into how we might get a short cable male/female DME connector for? I’m loathe to add complexity by adding another pair of connectors, but the idea of being able to log inputs while retaining the DME certainly has charms.

I wonder how many analog inputs my Traqmate can handle?

#45

I’m of the opinion that open vs closed tuning is not germane to this discussion. There are legal ways of tuning now (and illegal ways that are almost impossible to detect). Nor do I think that the idea of spending $2000-2500 for all new engine management stuff is a viable solution the the basic problem. Sure, it may work for a while but there is no assurance that some issue may not develop later. When (not if) it does one is faced with the problem of trying to figure out where the problem is. Something that could display live data or log data would save hours, days, weeks, or more of parts swapping. If a begginer has trolled this or other forums for a while I suspect they’d find the idea of an MS more palatible than the cost of an all new engine management system.

There will always be a perception that better hardware is going to make one faster. While there is some small truth in that, in a Spec E30 (or other cars for that matter) carrying 1mph more through a corner or eliminating a shift is going to be worth more than 10-15hp.

#46

the goddam germans got nothin to do with it! oh wait, yes they do.

#47

Scott,

I can take a look. I was more interested in the 25 pin PCB connector to put in an MS-II adapter board at the time, but I know the terminals can be had, so I’m sure the cable connectors can as well. We’d also need a little board to plug our logger into but that wouldn’t be a problem. It can be a temporary solution if you’re only interested in solving a specific problem & then going back to the simpler solution.

I’ve been looking at video/data logging systems & the comments about them on this forum. I’m sure the Traqmate & similar systems are excellent for learning to drive, but in my opinion they’re not really adequate for total health monitoring. Specifically, the Traqmate only has 4 analog inputs & 2 digital I/O, which isn’t going to go far enough to getting meaningful data unless you’re willing to pick only a few I/O to observe at a time. Plus you’d lose your shift light, oil pressure & temp, etc if you want to monitor engine parameters. I believe that 8 analog inputs is the practical minimum for such a system if you’re looking for a single logger solution.

The other issue is sample rate. I was told this weekend at the track (but didn’t verify) that it only samples the inputs & the accelerometer at 10 Hz (the same rate as the GPS chipset produces an NMEA sentence). I’m sure the ‘race’ loggers are similar in this respect. The GPS location data is only accurate to 3-10m, so the error in position data in that time is negligible but in 0.1 seconds, some of the engine parameters we’re interested in have changed dramatically. Specifically crank position signal at 6500 RPM sampled at 10Hz is going to be worse than useless. If this seems elementary, please understand that I’m trying to make it clear for the non-engineer. Of course, sifting through all the resulting data may be more technical than most want to handle and I wouldn’t blame them.

The only potential issue that I would point out is sampling sensors that are used by the DME & ensuring a reliable ground is maintained on the adapter board. If piggybacking of existing sensors is done incorrectly, you can affect the value that the DME measures, making it think a sensor is reporting a different value than it is, and clearly adding to your problems. This isn’t always trivial as evidenced by the army (2 or 3 to read essentially the same thing) of coolant temperature sensors on the M20B27 thermostat housing. One goes to the DME, one to the instrument cluster so as not to interfere with each other. However, most all of these issues have been solved before in the Megasquirt community and there is documentation on how to handle dual use of the same sensor, so we can apply the same approach successfully. Once the correct circuit values have been determined & verified, they can be used for all of our cars. Again, all of this could be built into a board on this adapter & then a cable run directly to the logger of your choice which would sample at an appropriate rate for each of the lines in question.

Phew. Let me look into the connector situation & we can discuss.

#48

You are correct, a little searching will show that Transtronics has a good high level article on how to adjust your tune using UV erasable EPROMs, a UV eraser, an EPROM burner, & either clone an existing chip, or tune your own. The gentleman at Atlantis Consulting appears (though I’ve never tried his software) to have already figured out where the Motronic 1.3 fuel & spark maps exist in the EPROM’s memory. These could be visually identical to the factory chip if someone was motivated enough to find the right parts. It’s more trouble than it’s worth (in my opinion) and there are ways to verify whether it’s been tampered with, but that’s also a lot of trouble.

That being said, I hope that this information is only used on non-spec cars …

#49

Whether MS is a good idea or not – I see little harm in letting Ranger (and hopefully a handful of others in different regions - if they’re willing) test it for a year to see if there is even any point in arguing over whether it’s good for troubleshooting/costs/the series/engine maps/etc or not. It may be the greatest thing since sliced bread and it may be just another costly headache that adds expense and difficulty. Without data - all we have is opinion. He’s offering to try it at his expense – what’s to lose?

#50

Matt from DIYAutoTune.com here. Ranger asked if I could check out this thread, so I’ve popped in to answer technical questions about the system. Ranger has proposed using a plug and play system similar to the Miata option we offer now, so installation would just be a matter of swapping out the stock DME, running a vacuum line, and loading it with a standard Spec E30 tune file. To answer the questions asked so far:

Probably $800-900 retail, and a group buy discount is a definite possibility. It will be based on newer and more capable technology than our current MSPNP line. That price would be pretty complete (including MAP sensor); you may need to add a USB adapter to connect to your laptop, but that’s it.

Yes, it can. While I would recommend using a MAP sensor built into the ECU, a MS system could be set up to use all the other original sensors easily enough. The MAP sensor would, as others have mentioned, probably be teed off the vacuum line.

#51

I see no harm in ranger trying testing it.

#52

Matt, glad to see you are here. Don’t forget about the temp sensor needed.

I have run the MS for 4+ years and other than to re-tune, never touched the unit. I built my own, which is an option for those so inclined. After Ranger establishes a tune, Matt could furnish that tune to anyone wanting to save the money and build their own.

One note…from all my tuning hours on the dyno, you want to setup the MS on a motor with the highest HP because that is the head that is flowing the most air…which needs the most fuel. Any motors making less power will be slightly richer so engine damage will not occur. Chuck

#53

[quote=“rsott” post=57882]He’s offering to try it at his expense – what’s to lose?[/quote]Have you seen the threads covering some of his other experiments? :woohoo: This ain’t gonna pass muster at any peer-reviewed automotive engineering journals.

It’s a new one car class: SERIOUSE30 - Spec Experimental Research, Investigation, Optimization, Undermining Simplicity E30.

#54

As they’re talking about keeping the AFM, you can use 100% stock sensors here.

Good point; this would be a good way to make sure the tune is fairly safe.

#55

I don’t know what I was thinking … it’s 35 pins for Motronic 1.0, and 55 for 1.3 … at any rate, it’s a Tyco (formerly Amp) Junior Power Timer connector, and the Tyco Electronics catalog page for the parts is http://www.te.com/catalog/bin/TE.Connect?C=17458&P=97869,14976,2697&M=COMP&LG=1&I=13&G=G&IDS=169873,170035

That page has the links to the relevant parts. I know some of the terminals are distributed by Mouser, you just have to look a bit. The connector housings could be sampled if you’re unable to get fewer than truckload at a time.

#56

[quote=“Herdrock” post=57888]I don’t know what I was thinking … it’s 35 pins for Motronic 1.0, and 55 for 1.3 … at any rate, it’s a Tyco (formerly Amp) Junior Power Timer connector, and the Tyco Electronics catalog page for the parts is http://www.te.com/catalog/bin/TE.Connect?C=17458&P=97869,14976,2697&M=COMP&LG=1&I=13&G=G&IDS=169873,170035

That page has the links to the relevant parts. I know some of the terminals are distributed by Mouser, you just have to look a bit. The connector housings could be sampled if you’re unable to get fewer than truckload at a time.[/quote]
Herdock, you’ve clearly got a great background on this, but go easy on the talk of unique-isms, complexity and tinkering. Our cars are supposed to be dirt simple and the same. Complexity makes folks nervous. Anything we do has to be simple to understand, easy to execute, (relatively) inexpensive, reliable, and pass a high bar of justification.

#57

I’m sorry if I’m running away with this. I want to make it absolutely clear that I’m not strongly in favor of any changes to the rules, just trying let folks know what’s possible. What you’re proposing is very interesting and I agree with the approach to testing but perhaps a logging only system can accomplish your goals without the need to be DQ’d in every race.

#58

[quote=“Ranger” post=57813]
Clean/flow injectors $130[/quote]
If you send them to RC Engineering to be rebuilt, then you are increasing your chances of at least one of them not working when you get them back (from personal experience). IMO, save your money…

Mark

#59

There are only so many M20B25 motors left in the world. 8 have already paid the ultimate price. How many more need suffer due to this new disease: Rangering. :laugh:

#60

Out of curiosity, how have these E30s managed to continue running for more than 20 years without using MegaSquirt ECUs? :unsure: While the OEM ECU will eventually fail somehow, aftermarket ECUs have a failure rate that is much higher than OEM units. So, if your MegaSquirt takes a dump at the track, where will you find another one? I hope you spent $500 to have a spare in your trailer, as compared to snagging an OEM spare in advance on eBay for less than $100 or trying to get one from a parts store (while the chances of that happening on a weekend may be slim, they are greater than rounding up a MegaSquirt).

How complex are these E30s that they need diagnostic logs, anyway?

IMO, I see no problem with installing a MegaSquirt in the car during your own non-event testing to find a problem, then reinstalling the OEM ECU for SE30 events. Testing the car at a local dragstrip or on a load-based dyno is much cheaper and easier than a wasted track weekend.

Mark (maybe building an SE30 one of these days, but working on Scott’s for now)