I'm in the process of upgrading from Mega-squirt II to Mega-squirt III.
First step is data collection, I all ready have a wide band O2 system. However I wanted to see how each cylinder is doing, since MSIII has the ability to tune each cylinder seperately!
The pictures are of a input/output extender and logger board module from JB performance. The module can be built to perform all most any function. Here's what I'm having perform
Monitor 8 EGT sensors
Monitor RPM and vacuum/boost
Monitor a knock signal from a GM knock sensor system
Monitor vehicle speed.
Monitor acelerometer reading from the built-in acelerometer
Future uses include boost solenoid control
Cooling fan control
Baro sensor readings for MSIII
and CAN comunication with MSII
The board is populated with the surface mount componets. Since the board can be used for many things and no two guys are going to use them for the same function. It is up to the user to install the misc resistors, diodes,etc.
The good news is Jean, the owner of the company who makes the I/O and logger boards, is very helpfull about instructing me how to assemble it!
While it is a do it yourself kit type of project. With help from Jean and others on the jbperf.com forums, it's coming along nicely.
input-outputandEGTloggerboeards.jpg
cuttingslotstheoldfashionway.jpg
finishedendplates.jpg
assembledboardsincase.jpg
The photos below show the assemble Input/Output extender, EGT logger module.
On one end is the slot for SD card recording and the slot for connecting a USB mini cord to a laptop.
The software used for viewing the data is Tuner Studio.
The other end shows the main harness attachment (DB37 connector) and the slot for where the EGT probe wires attach.
Also a view of the module installed under the dash on the drivers side.IMG_0895.JPG
IMG_0896.JPG
installedI-OEGTloggermodule.jpg
Pic's of the EGT sensors installed. Sensors are located 1 5/8" from the exhaust port. It is important for the sensors to be an equal distance from the port, for consistant temp readings.
The angle iron was used as a jig, to hold the sensor bungs in the correct location during welding.
IMG_0901.JPG
IMG_0902.JPG
jigforEGTbungplacement.jpg
Sequential fuel injection requires a cam sensor, to tell the ECM, is the cylinder is on a compression or exhaust stroke.
I used the top half of a Ford V6 cam sensor. Then machined it to fit on to the top of a shortened Buick 215 distributor.
The sensor is a single tooth VR type.camsensor.jpg
cansensorforsquentialinjection.jpg
Nice work Bill!
Makes me wonder if that I/O board could run my AA80E transmission.
JB
Thanks Jim
I would think that it could. Not sure how many shift solenoids the AA80E has. I know that the I/O board can be populated with up to 6 high current drivers. Most of the transmissions that I have had apart have between 3 to 5. Two or three shift solenoids, a TCC solenoid, and a pressure control solenoid.
Bill
Might need two boards then. It has 9 solenoids and most of them are PWM. Lots of inputs too. But that's another option for me to look into over the winter.
I really do like the work you've done on your car Bill. As one of the few LBCs with a blown and injected BOPR it's not just unique but a very sweet custom build as well. I don't think we hear from you nearly as often as some of us would like either. Keep up the good work!
JB
Bill, I don't have a clue as of what you do for a living but I bet that engineering comes to mind for anyone looking over your car. I've tried every possible way that I could think of to mount a blower on my 331 Ford small block but every supercharger I've tried were too wide to allow for enough room to install injector bongs. I had purchased a used B & M supercharger intake on ebay with the intent to do the same set up as yours; no need to say that the intake is now for sale.
I've also spent hours cruising the 5.0 Mustang web sites in search of a way to use the EEC-IV system with the Explorer DIS system (crankshaft fired). I did not want to use the ODB-II Explorer computer as we all know is tied to too many emission restrictions. Megasquirt was supposed to come out with a sequential system for Ford but it seems that it never materialized. I'm a firm believer in the superiority of the sequential injection versus bank injection. It has tested over and over to deliver better performance and MPG alike.
Anyway, I wanted to praise you in the cleanliness of your designs, it makes me drool every time that I look at the photos of your car.
Jacques
Thanks Jacques!
I agree about the squential injection, it has several advantages. Mainly I'm looking to improve in-town driveability and city fuel mileage. Not that its bad now, it's just that I work on new(er) cars all day. So I get to see the real advantage that the latest automotive computer controls has to offer!
I'm a tech at a GM dealership. I mostly do driveability and electrical work, so this is kinda right up my alley!
Bill
"Drivability tech", that explains a whole lot. In our shop our mechanics can advance/promote through ASE certifications. Out of 36 mechanics, only 6 have achieved that level. It is a very hard certification and I can honestly say that these 6 guys are our best mechanics. Some guys will be fore ever stuck on brakes and suspension as higher intellect are definitely required to achieve "Master Tech" and "Advance Drivability" status. Being 58 years old, I'm old school trying to catch up and being semi-retired it's hard to keep up with technology.
Since you work for a GM dealership, I assume that you're aware that the "Tech II" scanner is being phased out; no more upgrades will be offered and like Ford it is being moved to laptop software base. My company sends us to most new vehicle technology overviews that manufacturers offer; just like you, I love hearing about the new technology integrated in these vehicle. One example is when I found out that our K-9 Suburban had a camshaft phaser engineered in their 6.0 Liters, which quickly explained how their OHV engines are keep up with others in modern horsepower.
Do you have any ideas as of how I could combine the properties of the Explorer's (5.0L) DIS with the EEC-IV computer? The explorer uses the cam sensor in place of the distributor to announce the camshaft position. I would like to modify it for two reasons, (first) to achieve a zero ignition variation, (second) by eliminating the distributor, it frees up room in front of the engine different blower configurations and front intake inlet. Some guys would argue that it doesn't worth the headache but gear heads do a lot of things that fall short of practicality for bragging rights (LOL).
Good luck with your project and keep the photos coming.
Jacques
Jacques
It's been awhile since I played with EEC-IV! I assume that you want to use the Explorer's EDIS ignition with the pre-OBDII EEC-IV ECU?
You may be able to use a 95 4.6L EEC-IV ECU, since that engine also used EDIS. Although the timing curve in the ECU would probably be a little off for the 5.0L.
Best option might be to go with Mega-squirt III, then you can program your timing table. It also allows you to retard the timing by varying amounts, for a given boost pressure. Should be able to use the Explorer's cam sensor as well.
Yeah, the Tech 2 is slowly being phased out after all these years! Any new model coming out is using what we call GDS or Global Diagnosic System. It is indeed PC based, and I have a shinny new laptop to use at work! At this point the system is not "back dateable" to pre-GDS cars, so the good old Tech 2 will be with us for a little while anyway!
Bill
Bill, the reason I went with the EEC-IV (ODB-I) is that it escapes all of the changes that take place inside the computer after a code is set off like on the ODB-II. I've purchased a TWEECER "R" to change/manipulate timing and fuel tables on the OE ECU.
Megasquirt two years ago, was supposed to have a Ford version that would plug directly into the OE harness (ODB-I); everyone is still waiting for it. Anderson Racing makes a nice unit but a bit pricy.
The distributor provides two necessary signals to the computer for fuel and timing; there's actually a way to use the DIS but you lose the sequential fuel injection and turn it to a bank injection. Deep electronics understanding is my weakness; as you probably well know, pulling codes and diagnostics are one thing but understanding a system enough to modify it is another thing.
In another area, the EEC-IV was used for long time by the F-1 people with great success but that was long time ago. Now, if I could only find that EEC-IV Guru willing to help but that be too easy (LOL)
Jacques
Jacques
Looks like DIY has a plug and play Mega-Squirt for Ford EEC-IV. http://www.diyautotune.com/diypnp/models.html
Just another option, if the EEC-IV doesn't work out for you.
Bill
Thanks Bill, however after taking a look at the link, I still don't see where they offer the sequential function or the integration of the EDIS. If I can't find a decent system, I'm going to see how far I can push the techs at TWEECER-RT for fine tuning (LOL) My main desire is to recalibrate the timing and fuel curve to the vehicle weight and camshaft power band. It would be nice to run a distributorless system but I don't think that the money I'll have to spend justifies the small gain from it.
EDIS would be an interesting upgrade to modernize the car and eliminate fluctuation but it can always be done later. Please make sure to post the final results on your modifications, it always really keeps me curious/attentive to these great success stories.
Jacques
Jacques, I ran the EEC IV and EDIS on my engine for awhile (distributorless) (215 with blower) and used the Tweecer before removing the whole works and going to MS. My experience was that it was difficult to tune and had a great number of unexplained tuning parameters so a great deal of the time I was shooting in the dark. (Mustang SN-95 controller, before that Crown Vic). From the very beginning I had a mid-range surge and was never able to tune it out. In exasperation I finally ripped out the whole system and installed the MS. Immediately tuning and driveability improved and I never looked back. I don't think the loss of sequential made any noticeable difference. YMMV
JB
Doesn't the new MegaSquirt allow for sequential?
Dan B
Thanks guys for all the answers, I've followed Bill's link and used their MS message link to inquire about a plug and play that would carry EDIS and sequential injection; supposedly, it's now do-able, here's their reply:
Hello Jacques,
The basic plug and play setup is a bank to bank setup intended for TFI.
If you want to go with sequential fire and distributorless ignition, we
have another option. You would start with an MS3 and wire it to a DIYBOB
to make an adapter that is mostly plug and play:
http://www.diyautotune.com/diybob/index.html
This would take a bit of soldering, but is better suited to applications
that are going to need a bit of wiring regardless. Let me know what you
think.
Thanks,
Matt Cramer
DIYAutoTune.com Support
I have the I/O, logger boards up and running now. It's pretty cool! I can read live data, as well as data log 8 EGTs, MAP, RPM, vehicle speed and acelerometer readings!
I still need to do some more reading about using the acelerometer and calibrate it.
The EGTs are working nicely, and the temps are fairly close at cruise and WOT. There is some improvement to be had at idle and low throttle/decel engine loads. I should be able to dial those areas in when I switch to MSIII and fully sequential injection.
The parts arrived the other day to upgrade my MSII to MSIII/MSX and I hope to start on the ECU upgrade soon!
Attached is a couple pic's of my laptop displaying Tuner Studio with my custom I/O logger module gauges. Note: the EGT display is after I shut the engine off and the sensors were starting to cool down,
Bill
EGTguages.jpg
IOLoggerguages.jpg
That is just too cool Bill!
Did you have to get into much programming to get the gages you wanted and the datalogging you wanted? I never got past the standard displays and the spreadsheet datalog myself and basically gave up on it at that point. I could sure use some help on it next summer...
JB
Jim
Yeah it did take some figuring out! Fortunately the guys over at the JBperf forums are a big help.
Also my brother the electrical engineer was VERY helpful when it came time to edit the ini files for me! He understands the math needed to convert the raw data into usable readings much better then I do!!
Now that I have seen it done and played around with some I understand it much better now. I would be happy to help you out if I can.
Bill
Thanks Bill, you're a godsend. I probably won't get much done on it over the winter, and I realize this is bad timing, but in winter I do poorly on thought-intensive jobs like this. Still, just to get started on it, do you think you could send me the files you used to make your setup work? Maybe from that I can tell enough about it to make some progress, so that by next spring when I need it I might be a little more familiar with what I'm doing. I'd hate to have to do this kind of work for a living, it'd make me old before my time.
JB
No prob Jim, go ahead and PM me your e-mail address and I'll attach the modified ini file for the I/O Logger board.
I was trying to attach it to this post, however it wouldn't allow the file. Should work through e-mail ok though.
The way that I have it set up now, it even works as a stand alone EGT/MAP/RPM/VSS data-logger. It doesn't require a Mega-squirt ECU to function, just Tuner Studio down loaded into your computer.
Bill
Attached is a snap shot of Mega-log viewer. The top graph is RPM, MAP sensor, and Vehicle speed. Below are the 8 EGT readings. In this case, it was a free-rev in the shop, I hope to get some data-logs of the I/O Logger while driving as well. If it would ever stop raining, so I could get some traction:-)
The EGTs are closer together at higher RPM/engine loads, with them seperating under decell conditions. When I switch to MSIII I should be able to even the temps a little more under low load conditions, by utilizing the injector trim tables.
Bill
graph1.JPG
That's pretty cool. I sent Lance some of the info I dug up on the Lexus transmission I'm using and he's working on making the MegaShift (based on the General Purpose Input Output board) work with it, which would be great as it interfaces with the MS-II using CAN bus. That means a more complex GUI of course, so I've got my work cut out for me. I guess if I wanted to I could add a second GPIO for egt inputs like you have but it's enough of a challenge already. As it is I'm going to have to go to a digital speedo.
JB
Jim
I think once you go to a digital speedometer you'll like it. The VDO unit I have will read the VSS sensor signal straight from the sensor and is calibrated with a push of a button. It's the first acurate speedo I've ever had in a MG!
Attached is a couple more data-log snap-shots during a road test. The top graph is again MAP/boost, RPM and VSS, with the bottom three being EGT readings.
The EGT spread gets nice and tight together as engine load increases (boost conditions) then spreads somewhat when I let off the throttle. Also during light load/throttle cruise conditions. When I switch over to MSIII I will concentrate some of my tuning efforts in these areas to see if I can get the EGTs closer together. Although they are closer together then I thought they would be.
The MAP peaks are during a WOT aceleration run at 162 kpa or 9 psi of boost. It was interesting for me to see how quickly the supercharger builds boost at a fairly low rpm :-)
The bottom snap-shot is the same run only instead of VSS it shows the acelerometer reading (the squiggily red line). I will use this data to compare full throttle performance between the two different ECUs.
Bill
graphboost2.JPG
graphboostacelerometer1.JPG
I received the componets to upgrade my MSII to MSIII/MS3X.
Below are some more pics, one photo shows MSII daughter board on the left and the larger MSIII daughter board on the right. The daughter board contains the processor which is basically the "brains" of the ECU. The MSIII board also contains a SD card reader as well. The idea is to be able to data-log a road test to the SD card so that you don't always need to have a laptop connected if want record data.
Another photo shows the MS3X expander board. This board is used for fully sequential fuel injection, or it can be used for coil on plug (COP) ignition as well. The 3X board has 8 high output driver's, one for each injector and the injectors will be connected to the 3X board BD37 pin connector. Instead of the main MS mother board DB37 pin connector when "batch fire" injection is used.
Other planned mods to the MS mother board include, installing jumpers for CAN communication, adding a baro sensor for constant barometric correction, and adding a circuit for knock sensor input.
The black case is the new larger housing that is required to fit all this added "stuff" in to the box!
Bill
MSIIandMSIIIdaughterboards.jpg
Notice the MSIII board contains a much larger processor!
MS3expanderboard.jpg
The white ribbon strips are used to connect the 3X board to the MSIII daughter board.
IMG_0935.JPG
Nice looking hardware. I guess that explains what the guys have been working on.
JB
Jim M. and clan. I spoke with "Turbo City",(your picture).. They used GM's 2bbl/down drafts, rated 520 cfm@ 3"hg.87'-95'on sbc's.Used GM controller # 7747,re-programmed.Seems like(2) of these should, should feed a 5.7L hemi adequately, with Edlelbrock, dual plane, dual quad set-up. Anybody ? Thanks, Art.
Hey Art,
Check out the Chevy 454 truck/motorhome/marine TBI units.
Have also bored them out and used Holley spread bore butterflies. Gets you into the 690 range if you space the injectors.
Also check out Moates.net for chip burning stuff. You can make the GM ECU stand on its head and shoot watermellon seeds if you want to.
Cheers
Fred
Quite interestingly, I've been through (4) additional companies,since "Moates", regarding this seemingly impossible concept. No definative answers-yet. roverman.
Some more updates.
While the ECU was apart I decided to add some additional features. Such as a second MAP sensor for constant BARO correction, knock sensor input, and connected the CAN circuits so the ECU and EGT module can comunicate with each other.
The knock sensor system is a bit of an experiment on my part since knock sensors were never used on 215 Buicks. Knock sensors put out a specific frequency that has to be read by a knock module or the ECU. I thought about using Rover knock sensors, however they're signal is sent directly to the ECU. Since the Mega Squirt ECU cannot read a direct knock sensor signal, I decided to use GM knock sensors and a GM knock module.
With the GM knock system the sensor sends it's signal to a knock module. Then the knock module sends a simple 0-7volt signal to the ECU, which indiacates a "yes or no" for knock detected.
One of the main factors for knock frequency is bore size. So I decided to go with a 2.8 L V6 knock sensor(s) and module, because the 2.8L has the same bore size as a 215 V8. Not sure how the aluminum versus iron block will work out, but it may require some experimenting with different sensors!
I installed 2 knock sensors, one in each bank at the coolant drain plug fittings. Also I wired them seperately, so that I could dis-connect one or the other if it turns out that I'm getting a "false knock" signal.
Bill
The purple wires are the jumpers to connect the CAN comm to the ECU main connector.
The black pointer shows the "proto area" where additional circuits can be added by the user for additional features. In this case a zenier diode, resistors, and jumper wires were added for the BARO sensor and knock sensor input from a GM knock module.
IMG_0950.JPG
This view shows the two MAP sensors. The one on the right is the original sensor and is used as MAP sensor. The sensor on the left is the added in BARO sensor. Also shown is the 1/2 watt 1K ohmn resistor required for BARO input.
IMG_0951.JPG
Comparison view between the MSII case on the left and the new larger MSIII case on the right.
IMG_0954.JPG
Assembled MSIII unit.
IMG_0955.JPG
I have the car up and running on MSIII now:-) Still have some tuning to do (bad weather limits tuning days) however I have it very close already with just some fine tuning to do. Probably will play around with some of the settings just for the heck of it also!
Some observations so far.
1, The live tuning feature in tuner studio is really easy to use! Basically auto-tunes the VE table while you drive, which saves quite a bit of time when tuning the VE table.
2 Fully sequential injection has cleaned up the idle. Before when I had it batch fired I had to run a fairly rich idle mixture (around 13 to 1) in order to get a smooth idle. Now with sequential it idles smoothly at 14 to 1 for less exhaust odor! I suspect that it may get a little better fuel mileage as well. (haven't checked it yet)
3 VE table is adjustable to .1% instead of 1%. I noticed the most improvement in the 1500-2000 rpm range under very light throttle. Because of the very light load, this area is usually difficult to tune in that 1% up or down makes a pretty big difference in AFR. By being able to adjust a tenth of a percent at a time, its easier for auto-tune to dail into the correct air/fuel ratio.
4 Individual trim tables are working as advertised, using the EGT's I have been able to tune the different cylinders at various rpm/load levels. Still fine tuning this area, so far the results have been promising though.
Since all injectors can flow at slightly different rates at times, I'll probably send them in for cleaning and flow testing just so I know where each one is at for fuel flow. Then using the EGT data to "blue print" the injectors to the engine by installing the richer injectors in the leaner cylinders and visa-versa.
5 Auto-tune (closed loop operation) works much better (no surging) with the faster processor!
Overall I'm very please with the upgrade! While the MSII works pretty good, MSIII is definately better in almost every way. Kinda like comparing a computer built in the early to mid 90's to one that was made last year.
Bill
Good stuff Bill! It's great to hear the autotune works so well, I never had much success with it in the earlier units but I didn't give it that much of a chance. It'll likely be mid summer before my car is back on the road but I'm definitely going to take a good look at this setup. I have to add the GPIO board for transmission control and it might be time to upgrade the engine controller as well.
JB
Jim
Yeah I had limited sucess with auto-tune with the old ECU as well. It kinda worked, but in the end it was easier to just turn off auto-tune and tune it manually.
I suspect that it works so well now because of the faster processor. Also when it's doing it's thing, it is making the changes 1/10th of a percent at a time, instead of over-shooting and "hunting back and fourth" trying to find the correct air/fuel ratio.
I have CAN up and running between the two modules now. Up to this point I had been using USB to talk to the I/O board and using the standard DB 9 cable to talk to MSIII, then opening tuner studio twice in the lap top. With CAN I can read both modules through MSIII now, although I still need to change some of the MSIII CAN settings so that the EGT readings are the same as when viewing them directly from the I/O board though.
Bill
That sounds really good. If I do this right I'll have to run two GPIO boards, one like your setup and one for the transmission. At a couple bills each plus the MSIII that's beginning to get pricey. But, it should work the same as your system using the CANbus and both I/O boards should show up. As you're the trailblazer on that, did you have any sort of difficulties getting it to work that way? In the past the MS stuff has been pretty robust but I'll be wading in pretty deep this time and I'm a little anxious about that. Is there a particular person at JB Performance who is especially familiar with your system? It sounds like my best bet might be to duplicate what you've done and then add the 2nd extender board for the tranny. If there is an unused output on your GPIO that could also simplify things as I'm pushing it to get enough outputs for the tranny solenoids.
I was wondering though, and you probably already posted the info, what was the best thing you found for EGT probes and what did you use for bungs? Some sort of short thermocouple-in-a-stainless-tube? Those couldn't have been cheap either.
I've been sidetracked with building my office space to the point where I've gotten nearly nothing done on the car over the winter. It's about time to get off my butt and get it back in gear. Header mods are pretty high on the to-do list. Plus if I go this route that may push the completion date back to the fall. But I like the results you've gotten with autotune and think this approach has some real potential. Looks like I'd better finish cutting down the seat foams in Edith's TR7 so I'll have something decent to drive this summer.
JB
I did have several questions building and setting it all up. Jean the owner of JBperf is very helpful for when I couldn't firgure things out. He has a web site/forum just for his products http://forum.jbperf.com/index.php. Also there are other users who post helpful imformation on there as well.
For probes I used these http://www.thesensorconnection.net/egt-probes/all-egt-probes/316-exhaust-gas-temperature-egt-probe-with-compression-fitting Price wasn't to bad for EGTs, they provide technical data on them and they're American made! I know some guys have tried cheaper probes, but have had limited sucess with them ie inconsistant readings. I figured it's better to buy good ones, it's the whole "garbage in garbage out" thing! They came with screw-in SS bungs, I just ground the threads off and had them TIG welded in to the headers.
For the EGT's and I/O extender, I think I'm in it for just under $800. Spendy for sure, nice thing is that communicates with MSIII. Other 8 EGT systems I looked at were over $1000 and are stand alone only, where as the JBperf unit is both stand alone and CAN!
One of the side benefits of having CAN, is that MSIII can be programed to monitor EGT and if the exhaust temp reaches to high of a temp MSIII can add fuel to cool the combustion chambers down. It's an engine safety system that the tuner can use when he/she has EGTs. Looking though the settings the user sets up the perameters for exh temp, how much fuel is added,etc. I haven't activated this feature yet though.
Bill
Sounds good Bill, I have a single stand alone EGT on the opposite pipe from the wideband and used it to tune afr and timing quite effectively. Between the egt and the coolant temp I was able to get a real good idea of how far I could push it without adding too much heat. This would be even better.
Funny you know, this car was originally a very low budget performance car, and I mean low. You'd be amazed at how much money I didn't spend to do the original conversion back in the early 80's. The last few years though it's been getting downright spendy. So once I get it back on the road this time I think I'm going to just plan on driving the wheels off it. I guess that means I ought to spend what I have to to make it right, within reason, and this control system really looks to me like the way to go. I should be able to recover a little money by selling off the MSII systems I have now, that should ease the pain a little.
JB
Yeah, I was on the fence about whether to get a complete MSIII unit or get the parts/pieces to upgrade the MSII into a MSIII. I thought about selling the MSII or even keeping it for use in another car. In the end I decided to go the upgrade route, since I wasn't sure what I could sell the MSII for. It's probably pretty close to the same cost either way.
Bill
I'm not up to speed on the MSIII stuff yet Bill, as I was running MSII v 3 on both of my controllers. But it sounds like there may be some new features that make the newer one desirable and I would then have 2 extra controllers, both with the newer processor daughterboard, one of which has a knock sensor circuit. Probably have to stand on my head and rewire the main connector again. Well, I don't suppose it's a decision anyone can make for me and I'll just have to dive in and figure it out for myself, but it's not a process I especially enjoy so that and transmission control will very likely be the last step in the upgrade process and I'll put it off as long as I can.
I see where the MSIII has only been out for maybe a year. That explains why I didn't know anything about it. Last I heard the next new thing in the works was the UltraMegaSquirt which was supposed to have sequential plus stuff like ion based knock sensing. My impression of it was that it had sort of stalled out. So now there's MSIII which sounds maybe like another intermediate step? Also I'd noted that they had shifted a lot towards SMDs (surface mounted devices) which meant that building the boards was no longer an option and the prices were going up. Did that trend continue in the MSIII?
JB
Jim
You can still buy the MSIII as a kit, I think the daughter board and the optional 3X board are SMD only. Looks like the user can still build the main board though.
Not sure if they are looking past MSIII yet, it has more features then most guys would probably ever use. I have a feeling that they will stay with MSIII for awhile since it's so new.
With electronics you never know when it will stop, kind of boggles the mind:-)
Bill
It certainly does. What doe the 3X board do?
JB
The 3X board is for fully sequential injection, also will do COP ignition. Also has lots of extra inputs/outputs for all sorts of stuff! http://www.msextra.com/doc/ms3/hardware.html#wiring-ms3
About half way down the page is the externial wiring diagram for the 3X DB 37 connector.
Bill
OK, so you're running the MS3 controller plus the 3X board for sequential (plus it will run COP) plus another GPIO board for the EGT inputs, and then I'd need a second GPIO for transmission control. Does that sound about right? Then I see where your EGT probes cost $50 each. Did they give any quantity discount or did you have to pony up the full $400?
The tab for all that will add up, no doubt about that. But selling my two MS-II controllers should help. Luckily I have a full set of COP coils on the shelf if I decide to upgrade from EDIS. What did you have to change to get the MS3 to work with the MS2 wiring harness?
JB
Yes that is correct, no discount on the EGT's though (I asked). I wasn't sure if I wanted to spend that much coin for EGTs, but I figured with "non-standard" engine combinations such as ours, the information they provide is very useful. Makes it easier to optimize the combination, I'm hoping for a slight fuel mileage increase for when I'm just putting around town as well.
I haven't checked the fuel mileage yet, the car is on the hoist now. I'm fixing some oil leaks to get it ready for the drive to Reno this summer.
For wiring you can leave probably 90% of the wiring the same as it is now on your BD37. My old DB37 didn't have all of the terminals in it and I needed to add some wires for CAN. Since I had to replaced my main DB37 connector in order to add the CAN circuits, I went ahead and changed the rest of the pin-outs (mostly grounds) to match the MSIII diagram. Mainly just to simplify things.
Bill
OK that helps. Thanks Bill. Maybe I can do this in stages. My MG started out as a low budget build but it's been gradually getting ever more expensive. At some point I'll need to reign in the costs. But, as one of our friends is fond of saying, "It'd be a shame to spoil the car for want of a few dollars..." and we all know where that line of thought leads. I'll have to decide whether to get the controls or the sensors first. Maybe I should just install the bungs for now.
JB
I found out my existing MS v3.0 motherboard will work just fine with the MS3 daughterboard, which replaces the MS2 v3.01 daughterboard I have in the controller right now. That will give me the capabilities to run COP, sequential, EGT inputs, and datalog to a removable SD card. To get the COP and sequential outputs I'll need the MS3X extender board, and to use the EGT inputs I'll need either the JBPerformance IO-X card or a GPIO. Then I need a GPIO to run MegaShift to control the transmission, and it all gets linked together via the CANbus. Finally, I will need a new case large enough to house all of this. I should be able to sell my MS v2.2 motherboard and enclosure, as well as my other MS2 v3.01 card and my earlier MS2 v 2.0 card. My MS1 processor can go with the motherboard. So that should bring in a few bucks to help offset the cost a little. As soon as I understand the IO-X better the plan will be firmed up and I can start putting it into action.
Bill, I want to thank you for the friendly advice and assistance. You've made this process so much easier for me.
JB
Your welcome Jim! You'll have to keep us posted on how it's coming along for you.
Bill, did you go to COP? If you did, what did you use for coil drivers?
JB
Jim
I'm still running EDIS. I have a COP set-up from a 5.3L Chevy PU that I was thinking about using at one time. The EDIS has been trouble free for me, so I figured that I would stay with it for the time being.
Bill
I'm still sorting this out. If I use everything I'm considering it'll be pretty complex. But I have a new set of ACCEL COP coils that would look good alongside the Rover valve covers. They are 2 wire coils though so they need drivers, unlike the LS1 coils which are logic level and can be driven directly by the outputs of the MSX driver board.
My MS v3.0 mainboard will have several heat sink slots open after changing to MS3, perhaps 6 or more. So I'm looking at the possibility of adding coil drivers inside the case. I've begun a spreadsheet with pin assignments for the various boards, but there is now so much material out there that finding just the pin assignment information is not all that easy. Plus the MS3 hasn't been out long enough for the manual to be completed and the situation seems to be about the same for the IO-X board. I'll stay on it though and try to find a good place to upload the spreadsheet when it's finished.
JB
It's been awhile since I have posted so I figured it was time for an update. I'm very pleased with the MS3 and Jb perf logger board!
Tuning with the EGTs has been VERY helpfull, as it allows me to see whats going on at all rpm/engine load ranges.
A couple of things I've learned about tuning with EGT's.
1) Hotter temps usualy means leaner fuel mixture or retarded ignition timing.
Colder temps Usualy means richer fuel mixture or advanced/proper ignition timing (no detenation)
There are exceptions to the above however for example.
If the fuel mixture is to lean. A lean miss-fire can occur and actually drop the EGT reading, due to the un-burnt fuel mixture cooling down the exhaust.
If the air/fuel ratio is very rich, it can cause the EGT readings to go higher. This is because "the fire" is still burning fuel as it leaves the exhaust port.
If the EGT probes are placed to close to the exhaust port the readings with be hotter and visa-versa. The main thing is to have all of the probes equal distance from the exhaust ports.
Miss-fires can really throw off your readings much in the same way miss-fires effect oxygen sensors. Basically what a miss-fire does is make the oxygen sensor and EGT sensor read the opposite of whats really going on.
Since no two engines are going to run the same EGT temp at a given air/fuel ratio. It's very helpfull to have a good wide band O2 sensor set-up to know where your at air/fuel ratio wise.
For example, at WOT under 9 pounds of boost the EGT readings average around 1235-1250* F with an air fuel ratio of 12 to 1.
I also have seen the same 1235-1250* F readings at a lean higher rpm cruise with 14.7 to 15.7 air/fuel ratio. Same temp readings but at vastly different air/fuel ratio mixtures/engine loads.
Attach below are some screen shots of data logs.
The first one is a link to screen shot while the car was running on MS2
http://forum.britishv8.org/file.php?6,file=6916
MegaSquirt3EGTafterinjectortrimadjustment.JPG
Above is a snap shot after the switch to MS3 and after I did some tuning with the injector trim tables. Notice how much closer the EGT temp reading are. All 8 cylinders are there and they appear to be almost one line!
I had the ignition timing table set fairly conservative mainly because I didn't want to risk detonation. With an accelerometer (inside the JBperf module) and EGTs I decided to play around with the ignition timing some for a couple of reasons.
1) Using the accelerometer see if I could gain some more HP.
2) Using the EGTs see if the exhaust temps dropped when advancing the timing.
Below shows a snap shots during a 3000 4th gear low load lean cruise.
3200rpmcruisebeforetimingadvance.JPG
3000rpmcruiseafteradvancingtiming.JPG
Notice on the bottom graph the EGT temps dropped on average about 60*F, by simply advancing the timing 2* in this area. The idea is to slowly advance the timing a little at a time until there is no change in EGT temp. The bottom squiggly line is air/fuel ratio. For this tuing method to be acturate, the air fuel ratio must be the same in all test runs. I still may do some more timing tuning in this area, just to see if I can lower the temps more.
Now it was time to move on to some more WOT pull timing tuning. I wanted to be very carefull with this, because if there was detonation I would not be able to hear it over the blower whine!
Looking at the accelerometer data in the snap shots below, notice how the reading starts to trail off at about 4000 RPM. Most of this is because of the increased wind resistance (car is moving faster). However I had the timing pretty conservative in this area, I decided to leave the timing below 4000 RPM alone and advance it 2* at 4000 RPM with it slowly increasing by 4* at 6000 RPM.
The top graph is the before and the lower one after advancing the timing. Notice how the accelerometer reading stays higher longer. Additionally the seat of the pants acceleration felt stronger and the EGT readings dropped another 40* F as well.
WOTpullbeforetimingadvance.JPG
WOTpullaftertimingadvance.JPG
Looking good Bill. I'm still dealing with fabrication issues but all of this will be relevant soon enough. What are you using for an accelerometer?
I'm getting set up to coat my headers here. Went ahead and installed egt ports in each runner but they will probably be plugged initially, I still have the dash EGT with a probe in one collector.
JB
Jim
The accelerometer is a surface mounted unit, located in the JBperformance module. Accelerometer data can read directly from the JB board via a laptop, or through MS3 via the CAN communication line.
Bill
I had to do some fuel supply system repairs. On our way back from Reno we stopped at a park in Northern California, and my wife commented that she smelled gas. I smelled it too and got out to check the car over, to find fuel dripping from remote fuel cell at a rapid rate! After closer inspection I could see that it was leaking from one of the welds. Fortunately a fellow MG'er had some JB weld, so I pulled the tank, gobbed a bunch of JB weld over the welded area and let it sit over-night.
Got up the next morning and it wasn't leaking! I told my wife, we better head for home ASAP since I wasn't sure how long the JB Weld would hold. It was just starting to seep again by the time we got home.
Anyway I decided to eliminate the remote fuel cell and install the fuel pump in the tank.
IMG_1309.jpg
Donor fuel tank from out of a 98 oldsmobile. I picked this tank because of the nice flat area around the fuel pump mount. I cut out the fuel pump mount and soldered it to a stock MGB tank. Unfortunately I didn't take very many pics. It did require quite a bit of hammer/dolly work to get a nice fit on the MGB tank though. Due to the "raised areas " in the MGB tank.
IMG_1310.JPG
A 6" hole saw was used to cut the hole in the trunk floor.
IMG_1311.jpg
Stainless braided hoses and wiring in place. I reused my Astro van fuel pump, although I had to do quite a few modifications to it in order for it the compress short enough to fit in the MGB tank.
IMG_1312.JPG
Then welded up this cover out of sheet metal and covered it with carpet
Hers a pic with the carpet back in and everything all buttoned back up!
IMG_1313.JPG
A pic of my temporary roadside repair!
IMG_1314.jpg
I'm thinking about converting the tank into a tool box. It should make a nice heavy duty "battery box" tool storage area.
Bill