/\ / \ / /\ \ / ____ \ /_/ \_\
__ __ \ \ / / \ \ /\ / / \ \/ \/ / \ /\ / \/ \/
# ### ## # ## # # ### # #
_ _ | | | | | |_| | | _ | | | | | \_| |_/
####### ## ## ## ## ## ## #######
IMPORTANT: enter the case-INsensitive alphabetic (no numbers) code AND WRITE SOME SHORT summary of changes (below) if you are saving changes. (not required for previewing changes). Wiki-spamming is not tolerated, will be removed, so it does NOT even show up in history. Spammers go away now. Visit Preferences to set your user name Summary of change: '''Big Six Turbo''' BMW called M30 engines as "Big Six" (M20, just like mine is "Little Six"). This page is about an M30B34 turbo engine, originally in a 745i from 1985. http://www.bmwpower.hu/gallery/albums/userpics/10152/normal_45i%20017.jpg It's not a weak car even it's stock form (3.5 Turbo, automatic, see pic), but this engine is now in a light chassis (e30): http://www.bmwpower.hu/gallery/albums/userpics/10152/normal_alkatr%E9szek%20022%7E0.jpg This page is about GenBoard-ing it! '''Original setup''' Car outlines: * Engine: M30B34 turbo, 3400cc, '''6 cyl inline''', 12 Valves, SOHC * Fuel injection: Bosch injectors (part no. 0 280 150 200) maybe 270-300 cc/min @3/3.5bars (different sources say different values). R=~3 Ohms. * Fuel pump: Stock * Stock 3 bar pressure regulator, with manifold pressure reference * '''Turbo''': stock KKK K27 * Idle control : stock. (similar to M20) * Intake: stock AFM, Small stock air-to-air intercooler * Factory throttle pos switch (3 state; idle, between and WOT) * VR crank-trigger: 60-2. ** M20 baby series engine uses exactly the same crank triggering, see MembersPage/GergelyLezsak/Trigger * Cam position sensor - coil at plug wire #6 * Coolant sensors - similar to M20 * '''Knock'''. Two factory installed knock sensors. * More sensors: O2 sensor: narrowband , IAT in AFM * orig '''ECU''': Bosch Motronic 0 046 430 136 0 Buylist at MembersPage/GergelyLezsak/BuyList ---- '''Project''' Electronically the M30 engine seems very similar to the M20 little brother (from the same era: motronic 1.3). Even their original ECU can be swapped, and the engine will start up! (mapping is not the same, so don't use this as a common practice) After i've received[http://quasar.dynaweb.hu/~lezsi/bmw/conversion/345i/parts-1.jpg first part of the order] I made a new harness, and ordered a new 3.3 GenBoard. Now, the car is running stable with GenBoard. ---- '''UPDATE''' '''2007.01.07''' The car's got new 500cc injectors now. I've replaced it's original GenBoard with a new one, as we couldn't find the reason for silly MAP values for a long time now. (External sensor were replaced, MAP +5V voltage is measured with DVM, still got much lower values than reality -but only when the engine is working.) Unfortunately we couldn't map the car with the new injectors. Details at MembersPage/GergelyLezsak/BigSixTurbo/InjectorsTwo '''2006.07.25''' The car is stable now, had an 1/4 mile drag strip with 12.88s best time. Stock 300cc injectors were not able to keep up high rpm @high boost (up to 1.4 bars). Had some problems with ignition coils (this is the second one which gives the spark across its housing with high humidity -the newer type with plastic housing). Interesting: there's a strange behavior of LCD when spark is in the engine compartment: the characters are falling apart; sometimes it's a bit misaligned but sometimes totally unreadable. If you see something similar check your ignition system! :) There's no change to silly map values, but we'll settle with it for a while, because behavior, power, fuel consumption is acceptable. Soon we need to upgrade injectors and then remap VE, so that's a good time to cure this problem also. '''2006.05.16''' Some news. We had serious problems with the car's power and consumption, it turned out to be an ignition retard problem, which was caused by new ALS function accidentally enabled. (AFAIK newest MegaTune doesn't show the config variable '''ign_balance''' which was FF and caused the trouble. Thanks for Marcell's support!) After setting it right, ign. adv. get normal and the car gained ~100HP, +50% fuel economy :) -Interesetingly boost pressure falled back (from 1.2/1.5bar to 0.9-1.2bar)with good ignition and power. We didn't deal with knock and boost control this time, since the car dealed with the turbo's max boost level well. But MAP values are still silly. I checked values with engine not running. At ambient air pressure it's fine ~100kPa, with connecting GND to MAP input shows 000kPa, Connecting +5V (which was actually 4.93V) to MAP input shows 249kPa. With the car running it showed inaccurate values like before. Replaced the MAP sensor to another similar one, but values remained the same. When engine running I see ~70kPa at normal ambient pressure (0 bar on analog boost gauge, double checked). When engine is off, MAP goes back to 100kPa@ambient pressure. I think two sensors can't be wrong similarly, so I consider MAP sensors good. Displayed values at 0/100/250kPa are also fine so I guess analog input->GenBoard internals->LCD with firmware are also OK. What else could be wrong? Maybe +5V reference voltage is changing while engine running? Any other idea? *Just 1 possible thought, what is the Map sensor from? I had similar problems with incorrect map readings when I was trying to use a Honda map sensor. It still had a linear rate but read way different than what was actually there. I was able to actually tune the engine still but it was hard to think of idle as 70 kpa so I replaced it with a GM type sensor and all was bang on. EricN * Thanks Eric, these are sensors from webshop, and I think it can't be the range/slope problem because of the 70/100kPa shown deviance of ambient pressure, depending on the engine's state. Definitely we could measure voltages during travel, and/or testing more pressure reference points, but that's need more preparation :( '''2006.05.12''' After a lot of mechanical failures (cracked head, etc), car's running again, but mapping is still poor. I'll definitely look for correct MAP values first (still not investigated). . . . '''2006.04.03''' At least we changed to the latest firmware. (1.0.38) BMW-style dual IAC valve is fine now. And car had been mapped nice. One strange problem with MAP: on boost, MAP values are less than real (probably the half of real) values. * In vacuum it seems OK: decceleration: 15-20kPa * idle: 45-55kPa * '''free air: ~100kPa''' * But like 1-1.2 bar boost showed only 150-160kPa MAP instead of 200..220kPa * 0.2 bar boost showed like 95kPa MAP instead of 120 kPa. (lower pressure than free air ????) I'm using 250kPa MAP sensor, and double checked the values with two (known good) analogue gauges. I probably misconfigured something with the new firmware. Config and tables are here: http://quasar.dynaweb.hu/~lezsi/bmw/attila/configtables20060402.txt By quick look, I couldn't spot anything to justify bad MAP reading. Preferrably with pressurized MAP connection (eg. using a compressor) Do the voltage measurements, or at least read the sensor_filt[] with TerminalProgram in hex-view mode with mdx23 command. mdx.. like mdx23 spits out 32 bytes from 0x230 SRAM address: since sensor_filt[] is a little-endian array of 11 bit sensor ADC readings starting from 0x23f in the 1.0.38 firmware, positions in adc.h 2=>TPS, 3=>CLT, 4=>MAT, 5=>MAP so MAP will be at 0x23f + 5*2 = 0x249 showing 0xff 0x07 (little endian 2047) when MAP input is close to 5V After this, we will know where to look: in hardware (eg. MAP connection, supply) or firmware (eg. configuration). '''In any case, check/measure the GND connections very carefully !!!''' '''Needed:''' * mcd ** most importantly kpa_conf ** kpafac ** kpa_offs * mct * if you can hook up an external pressure sensor, (or a DVM if you alredy have an external) that could also provide verification info '''2005.12.28-29''' Car is running with GenBoard :) We'have removed the whole factory harness. (Took a while to reproduce everything) Some experience: TPS wire's and IAC open/close wire's direction is always replaced :) Had some difficulties with digital outs; EC36pin15 (Which is P259/5) didn't do anything (with the fuelpump). Tried to move it to the next pin (P259/6), resulted the same. Finally chose a different chip: S259/5, EC18pin5. It's working. Couldn't imagine what could be wrong, at the test time fuelpump out was working. After we've got fuel I made it flooded. I don't know how but my previous calculations worthed nothing. I thought fuel_req at the 12ms range, but I needed to half that, and maybe we'll half it again. Idle (which is horrible now, but that's mainly pid-related) is around 1ms pw. Don't know why, these are ~300ccm injectors. Maybe too slow to close? Will try to lower pwm value of 30%, and PowerFlyback if necessary... Since we've got 25cm of snow the day the car started, couldn't tune it yet... Some pics: Dashboard: http://www.bmwpower.hu/gallery/albums/userpics/10152/thumbnails_analog%20muszerek%20010.jpg [http://www.bmwpower.hu/gallery/albums/userpics/10152/analog%20muszerek%20010.jpg in big] Nice work with the LCD mounting. The first day at the backyard (no snow at all): http://www.bmwpower.hu/gallery/albums/userpics/10152/thumbnails_udvar2.jpg [http://www.bmwpower.hu/gallery/albums/userpics/10152/udvar.jpg in big] ---- '''TODO''' * LCD with cable connection ** Done: not exactly with the specified resistors, but it's perfect. [http://quasar.dynaweb.hu/~lezsi/bmw/conversion/345i/lcd1.jpg pic1], [http://quasar.dynaweb.hu/~lezsi/bmw/conversion/345i/lcd2.jpg pic2] * Make whole harness ** Done * External MAP housing with connections ** Done * When GenBoard is ready, flash it with my IAC-tweaked firmware ** Done. It's included (and fixed) in latest - post 1.0.32(?) firmware. * Mapping ** It's going to be a whole new story at MembersPage/GergelyLezsak/BigSixTurbo/Injectors * Setting IAC table/PID ** It's still an issue for this and my own car. Valve control seems allright, but can't get through the problems of very low RPMs (and sometimes stalling) at sudden throttle release. Details soon. * Boost control plans - MembersPage/GergelyLezsak/BigSixTurbo/BoostControl * Calibrate and test knock sensors. Checked dox but can't find out what knock1_frequency values actually means? ** the value directly from TPIC8101 datasheet (table), usually 0x29=41 but check for your freq. What is your freq? *** That's it! We don't know the freq yet but now I understand what I'll adjust, fine! :) ** knock plans - MembersPage/GergelyLezsak/BigSixTurbo/KnockSense * PowerFlyback? - Sure!, discussed at MembersPage/GergelyLezsak/BigSixTurbo/Injectors page ---- * Here is a similar car: MembersPage/MattiasSandgren/BmwSevenTwentyEight - Unfortunately the owner had to sell it for 1300 euro.. * Another similar car: MembersPage/MattiasSandgren/BmwFiveTwentyEight (Will be my daily driver //Mattias) and a [http://media.savarturbo.se/~mattias/vems/bmw_e23_728_turbo/pimpride_2005.avi movie clip] for motivation //Emil Larsson ---- Optional: Add document to category: Wiki formatting: * is Bullet list ** Bullet list subentry ... '''Bold''', ---- is horizontal ruler, <code> preformatted text... </code> See wiki editing HELP for tables and other formatting tips and tricks.