This project is in a weird state. Do not experiment without a concept (the most likely outcome is wasted time and perhaps damage.
Do not experiment with known trashed config. It does not make sense.
Pointing to a wiring that apparently does not even match the connector type => unlikely that it could work this way.
I purchased a 2nd hand home-assembled PnP? VEMS v3.3 and have installed it on my Mk1 VW Golf 2.0 ABF 16v Turbo. The former owner has written a lot of info about the assembly here:
- PNP means that (according to referred page) your frontplate has T68 main connector (non-factory assembled device).
- but this is inconsistent with the controller you sent in, which does NOT have T68 connector, but standard econoseal EC36+EC18.
- Jason suggests this might be an "adapter harness" (not PNP). Can you take a photo (near the ECU connector) or use a better description ?
- THANKS FOR THE UPDATE - MUCH APPRECIATED: You're right, PnP? is the wrong term, it is an adapter harness where the original harness's multiplug pinouts have been used and then run to the EC36 and 18 connectors.
- in any case, watch the ground rules, easy to get wrong with a homebuilt PNP device. Thanks - all grounds checked for consistency and are fine.
The v3 #671 mainboard was replaced after damage, to brand new #1746. Hopefully the new is not killed yet. Appears to be perfect.
It would be best to write your own pages. Fine to start by copying parts from Basil (0 time), but make sure to only include the wires that you really have, and verified, preferrably end-2-end from engine connector to ECU connector. Including the flyback and the 5 ground wires!
You can update your own page any time (you understandably don't change Basil's), and to get things sorted instead of getting them tangled more and more.
- UPDATE 2/July/2008
Rob from VEMS UK sent me a known good config, which can be viewed here:
The crank signal is now perfect and everything else appears ok
- but only have spark to cylinders 1 & 4
- see Ignition outputs below. You apparently wired ignch 0 and 1, configure accordingly
The car runs wastespark Bosch VW coil (exact type or photo maybe?) with the factory ignitors removed.
The coil is identifiable in this picture:
It is used on all Golf 4s as far as I am aware.
This is the complete engine:
Checklist - unlikely to be a problem though
Only after you stop experimenting with known trashed config. First upload a known good config and check these if you still have a problem.
- I don't have a copy of the original config as I never generated it when the VEMS first arrived
- is the input type jumper set OK ?
- Trying a VR input setup inside the box, with a HALL sensor connected could give trigger problems
- MembersPage/BasilDaham/CordobaSX lists HALL, and we are trying a HALL sensor, but ...
- theoretically HALL, HALL trigger, but it could be a good idea to doublecheck the blobs near the LM1815 (the one closer to atmega), to see which one is shorted ( GenBoard/Manual/InputTriggerHardWare and GenBoard/VerThree/Schematic ). (can be seen with removing endplate only, do not unclamp the device)
- is the wheel the same 60-2 HALL type as earlie ?
- VR wheel is not stable for HALL
When we first installed the computer, the car ran ok, but with a weak spark. After a few starts, the crank signal started to become intermittent: it would appear and then disappear and the car would run and then fail to run without the signal.
Would be typical with insufficient grounding, but we know little about wiring so hard to tell.
Eventually, the RPM signal disappeared entirely and "trigger errors" showed up on Megatune.
- After checking all the grounds (all fine)
- this may contradict the other info (down): remade all of the grounds => RPM returned
- and everything else, we decided it must be a hardware problem, so sent the ECU back to Hungary where said it showed signs of damage and replaced the board.
- with the new mainboard, first had no crank signal.
- pulled harness out and remade all of the grounds to make entirely sure they were ok. After doing this and replacing the crank angle sensor (just to make extra sure) the RPM signal returned.
- However, the ECU had come back with the 1.0.73 firmware installed and this resulted in no spark.
- Not being entirely sure how to fix it with all the config. stuff
- STOP. Do not start the engine, before wiring details are clarified, firmware and config uploaded.
- STOP experimenting, start to detail the wiring, measure while you're doing it
Since this point, the RPM signal has become intermittent again:
- It jumps to 250rpm under cranking and back to 0rpm all the time: signal, no signal, signal, no signal
not surprising: if suitable config not uploaded
- you likely not have tooth_wheel=3A
- in MT primary trigger settings number of tooth on wheel=58
To untangle this, subsystems should be moved (deleted from this page) to their own subpages. Only leave uptodate info there. If writing any info that is old (eg. config dump read BEFORE the most recent firmware upgrade) :
- MembersPage/StuFrench/TriggerSetup (factory multitooth 60-2 HALL on crankshaft)
- there might be some internal trigger setup uncertainty, although it should theoretically be HALL, HALL
- the controller was admittedly damaged due to bad install (missing flyback) earlier, but hopefully the wiring was fixed.
- The ground was fixed
- the flyback was also fixed ?
Advanced filtering: Disabled
Special N-1: Disabled
Fiat Stilo: Disabled
TDC after trigger: 60
Number of teeth on wheel: 58 are you sure ?
Trigger tooth: 3
Next trigger tooth: 30
Crank min period: 3744 usec
Angular width of tooth: 2
Tooth relation Normal: 122 (irrelevant with multitooth advanced filter disabled)
Tooth relation Missing: 181 (irrelevant with multitooth advanced filter disabled)
Angular width of missing teeth: 6
Engine phase: 240
CRANK SENSOR WIRING:
- EC36pin28 for 5v supply
- EC36pin27 for signal
- Sensor ground to EC36pin26
I AM CONVINCED THIS MUST BE A CONFIG PROBLEM
- the crank sensor is performing fine and the car HAS run before.
- important: When I first got the car running upon installing the VEMS and harness, I opened Megatune using Megasquirt version of the program, and clicked 'ok' when it warned that table corruption may occur. The car has never run properly since this point in time.
- likely you just need to upload the config and tables, it's absolutely pointless to experiment with a KNOWN TRASHED config
- to verify, dump with TerminalProgram mcd / mct,
- or more userfriendly upload tool: recent [vemstune] to download config (you can also upload with it as well)
- than compare to your known good config.txt
- At this stage, this is my saved config.txt for 1.0.73Final
- what is "this" stage ? Is it dumped when the engine was working or after the whole config was THRASHED ?
- Tables can be checked against MembersPage/BasilDaham/CordobaSX info
Ignition outputs with spark table H2 values changed to from 1.0.53 to suit 1.0.7x:
If you have i259/1 and i259/2 than
- from h=01 02 01 02 01 02 01 02 (1.0.5x firmware)
- to h=10 20 10 20 10 20 10 20 (1.0.7x firmware)
But MembersPage/BasilDaham/CordobaSX suggests i259/1 and i259/0 like maybe
- cyl2,3 EC36-35||I259_0||
- cyl1,4 EC36-33||I259_1||
h=10 00 10 00 10 00 10 00
j=6580 6580 6580 6680 6780 6880 6980 6980 6B80 6980 6280 5B80
j=6480 5D80 5B80 5F80 6580 6880 6980 6880 6C80 6D80 6780 5E80
j=6380 5B80 5980 5E80 6380 6780 6980 6880 6D80 6D80 6980 6180
j=6380 5980 5680 5980 6180 6480 6680 6680 6B80 6F80 6B80 6380
j=6380 5980 5880 5880 5F80 6280 6580 6580 6A80 6F80 6D80 6480
j=6480 5F80 5A80 5A80 5E80 6180 6380 6580 6A80 6F80 6E80 6780
j=5F80 5F80 5C80 5B80 6080 6380 6480 6780 6B80 7180 7080 6980
j=6580 6280 5C80 5C80 6080 6480 6880 6C80 7380 7680 7580 6C80
j=6F80 6B80 6880 6380 6380 6780 6B80 6F80 7780 7A80 7980 7080
j=7880 7680 6E80 6A80 6880 6B80 7080 7680 7C80 8080 8080 7880
j[A]=8580 8380 7F80 7B80 7A80 7D80 8180 8680 8B80 8E80 8C80 8480
j[B]=8E80 8D80 8B80 8880 8780 8A80 8D80 8E80 9380 9680 9580 8F80
l=3D 36 2F 2F 2F 2F 2F 2F 2F 2F 2F 2F
l=37 34 31 31 31 31 31 31 31 31 31 31
l=37 36 31 31 31 31 31 31 31 31 31 31
l=37 37 37 37 37 37 37 37 37 37 37 37
l=37 37 43 47 4F 4F 4F 4F 4F 4F 4F 4F
l=47 47 4F 55 58 58 58 58 58 58 58 58
l=66 66 66 66 66 66 66 66 66 66 66 66
l=6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B
l=6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B
l=6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B
l[A]=6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B
l[B]=6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B 6B
n=D0 08 08 30 68 A0 BC BC BC C0 C0 C4
n=40 08 08 34 68 98 BC BC BC C0 C0 C4
n=40 08 08 38 70 94 BC BC BC C0 C0 C4
n=40 14 14 40 78 94 A8 B0 B0 B8 B8 C0
n=40 1C 1C 4C 7C 94 94 A0 9F A8 B0 B8
n=2C 2B 2B 4C 78 8C 94 88 88 90 94 9C
n=2C 38 38 48 70 74 78 70 70 78 7C 7C
n=2C 38 38 44 68 6C 70 68 68 70 74 74
n=2C 38 38 3B 54 58 5C 54 53 5C 60 64
n=2C 30 30 34 48 48 4C 48 44 48 4C 54
n[A]=2C 30 30 2B 30 30 30 28 28 2C 30 30
n[B]=2C 1C 1C 18 10 10 0C 08 08 0C 10 10
k=0A 0F 14 1C 22 28 2B 3C 4B 5F 78 96
r=07 08 0C 11 18 1E 23 28 32 3C 46 52
h=02 08 10 41 00 00 00 80
h=00 00 00 00 00 00 00 00
h=10 20 10 20 10 20 10 20
b=14 18 1C 20 24 28 2C 30 80 80 80 80
t=21 40 60 80 A0 C0 E0 FF
- I know how to edit, save and upload the tables, but don't understand them
Are you absolutely sure you uploaded this config and tables ? It sounds like you uploaded firmware without uploading config + tables.
That means the above mcd/mct might not be what you have in. You can download, verify again.
Or you can just do what should have been essential in the first place, and upload your known good config.
If still no success
publish details about your setup (see IssueReports for checklist )
- mcd / mct dump ( vemstune, "config download" )
- not the config that you certainly know is not in there anymore because it was trashed, but the downloaded mcd / mct. This is very important.
- Looking at a config that has nothing to do with what you have in now is pointless !