Help Distributor 3 pin plug

[gajh]

No not at all I am just glad of the help, once i get the replacement connector and redo the wires I am hoping the car will work a bit better. I have seen some documentation that states if you unplug the hall sender then the DME retards the ignition time by 4 or 6 degrees BTDC. I am not a mechanic but it sounds bad so best get it sorted

 

[barks944]

Yes, I think the reason it has the hall sensor is so it can work out which cylinder is knocking (if there is one) this is used to individually retard the ignition on a cylinder. When it doesn't have this information it seems like it defaults into a safe mode to protect the engine.

Ignition timing is very critical in producing power as you need to time the ignition at the right point to get peak cylinder pressure at the correct time. Retarding the ignition by 4-6 degrees will cause a large loss in engine power but it will prevent knock and protect the engine.

 

[gajh]

thanks for that a very informative reply, cant wait to get all the work done so i can drive the porsche and sell my alfa

 

[gajh]

one final question chaps, can i remove the rear and front cam cover (i think they magnesium or puter) with out disturbing the belts, was think it is looking a bit of a mess so thought i could powder coat this while I am at at it. Thanks

 

[bertelli_1]

ORIGINAL: barks944

Yes, I think the reason it has the hall sensor is so it can work out which cylinder is knocking (if there is one) this is used to individually retard the ignition on a cylinder. When it doesn't have this information it seems like it defaults into a safe mode to protect the engine.

The bosch system on the '44 isn't quite that advanced, it cannot adjust the timing of individual cylinders (this is something that you only see on recent high end stuff). The knock control is not linked to the hall sender either.

The rear cambelt cover can't be removed without removing the cam pulley (and belt) first I'm afraid.

 

[gajh]

thanks for the reply, i will just have to try clean it and tidy it up while still in the car

 

[barks944]

ORIGINAL: bertelli_1

ORIGINAL: barks944

Yes, I think the reason it has the hall sensor is so it can work out which cylinder is knocking (if there is one) this is used to individually retard the ignition on a cylinder. When it doesn't have this information it seems like it defaults into a safe mode to protect the engine.

The bosch system on the '44 isn't quite that advanced, it cannot adjust the timing of individual cylinders (this is something that you only see on recent high end stuff). The knock control is not linked to the hall sender either.

The rear cambelt cover can't be removed without removing the cam pulley (and belt) first I'm afraid.

Do you know what exactly its used for? On an injection system with control over individual spark plugs/injectors cam info is clearly important so you know what cylinder is on what stage of the 4-stroke cycle. I'm not sure what the S2's DME would use it for as the Dizzy takes care of the which cylinder sparks and the injectors (at least on the lux and turbo) batch fire.

 

[barks944]

Sorry to go a bit of topic, but I notice in your website bertelli_1 that you had your clutch refurbished. Who did you get to do this? Nice project BTW keep up the good work!

 

[bertelli_1]

ORIGINAL: barks944

ORIGINAL: bertelli_1

ORIGINAL: barks944

Yes, I think the reason it has the hall sensor is so it can work out which cylinder is knocking (if there is one) this is used to individually retard the ignition on a cylinder. When it doesn't have this information it seems like it defaults into a safe mode to protect the engine.

The bosch system on the '44 isn't quite that advanced, it cannot adjust the timing of individual cylinders (this is something that you only see on recent high end stuff). The knock control is not linked to the hall sender either.

The rear cambelt cover can't be removed without removing the cam pulley (and belt) first I'm afraid.

Do you know what exactly its used for? On an injection system with control over individual spark plugs/injectors cam info is clearly important so you know what cylinder is on what stage of the 4-stroke cycle. I'm not sure what the S2's DME would use it for as the Dizzy takes care of the which cylinder sparks and the injectors (at least on the lux and turbo) batch fire.

On the turbo the injectors are fired in pairs (or semi sequentially as some like to say - unless you know different from your recent work??), on the s and s2 they are batch fired (all 4 together). The s and s2 only have 1 crank sensor, the crank trigger is a common bosch 60-2 wheel. On cars fitted with the 60-2 trigger (60 teeth, 2 missing, still used on most vag cars) the ecu can work out the tdc position from the missing teeth. The cam sensor is (probably) used to trigger the ignition amplifier (via the ecu) to control ignition timing - this would explain why it defaults to 6 degrees without a signal - as suggested earlier. I haven't experimented so I can't confirm that.


eta: if you want to see some clever knock / spark timing stuff do a google search for saab ion sensing.

 

[peanut]

ORIGINAL: bertelli_1


On the turbo the injectors are fired in pairs (or semi sequentially as some like to say - unless you know different from your recent work??), on the s and s2 they are batch fired (all 4 together).

I understood that the late turbo fuel delivery system was the same as the S2 (Motronic )the injectors are fired simultaneously once per crankshaft revolution (or twice per four stroke cycle)

The injectors are constantly live and are pulsed grounded within the ECU .If you look at the internal ECU schematic the injectors are wired in pairs but I thought they are all fired simultaneously on the late turbo ?

Bosch certainly made continuous modification and inprovements to the Motronic system over the 89 -92 period so you may well be right. It would certainly make sense to deliver half the fuel every crank revolution rather than have fuel hanging about in the inlet chamber for the next crank cycle

 

[bertelli_1]

ORIGINAL: peanut

ORIGINAL: bertelli_1


On the turbo the injectors are fired in pairs (or semi sequentially as some like to say - unless you know different from your recent work??), on the s and s2 they are batch fired (all 4 together).

I understood that the late turbo fuel delivery system was the same as the  S2 (Motronic )the injectors are fired simultaneously once per crankshaft revolution (or twice per four stroke cycle)

The injectors are constantly live and are pulsed grounded within the ECU .If you look at the internal ECU schematic  the injectors are wired in pairs but I thought they  are all fired simultaneously on the late turbo ?

Bosch certainly made continuous modification and inprovements to the Motronic system over the 89 -92 period so you may well be right. It would certainly make sense to deliver half the fuel every crank revolution rather than have fuel hanging about in the inlet chamber for the next crank cycle

That all sounds plausible. The odd thing is they way the injectors are paired, 1&2 and 3&4. You would expect 1&3 and 2&4.

 

[barks944]

On the Lux and turbo there are two outputs from the DME to two pairs of injectors. However these two outputs are both driven by a single output from the computer in the DME. All four injectors fire together.

The way the motronic was built was to make it very configurable, there are many features in DME x that were not put in DME y etc etc. They probably designed the connector and loom to support some sequential injection on car X but it wasn't done on the 944.

http://www.youtube.com/watch?v=pp7hOKCuK8Y

Just reailsed this video doesn't show all four injectors! But they were all firing off a single microprocessor output.

The 944's injection system does inject into a closed port. In fact 75% of the fuel injected by the 944 is into a closed port. Theres simlpy no way round this with the design of the system. In fact most cars do this, albeit at high rpm's when the time where the intake valve is open is too short for the fuel system to deliver enough fuel at high engine power.

 

[peanut]

ORIGINAL: barks944

On the Lux and turbo there are two outputs from the DME to two pairs of injectors. However these two outputs are both driven by a single output from the computer in the DME. All four injectors fire together.

The way the motronic was built was to make it very configurable, there are many features in DME x that were not put in DME y etc etc. They probably designed the connector and loom to support some sequential injection on car X but it wasn't done on the 944.


Theres simlpy no way round this with the design of the system. In fact most cars do this, albeit at high rpm's when the time where the intake valve is open is too short for the fuel system to deliver enough fuel at high engine power.

we were discussing the late turbo and 16v cars specifically . The Lux and early cars have a different ECU and different sensor and fuel injection and ignition control to the later 16 valve cars.

As you say there are some frustratingly obscure minor differences between models as the 944 Motronic system was modified through to the Cabrio in 1992 .

With regard to there being 'no way round the pulsed fuel system' there is of course the 'continuous fuel injection' systems which meters air and fuel and delivers the mixture continuously .The 1974 911Turbo used this system as did a lot of the 1986 era Mercs. Easily identified by seperate fuel lines feeding each cyclinder as apposed to a common fuel rail.

The Bosch fuel injection and engine mangagement system makes a fascinating study (except when you have got a non starting car )

Hows the DME project coming along ? About time someone did this. I intended to design and manufacture a fuse/relay board for the 928 a few years back but sold my 928 before I got started fortunately []

ps if you fancy making your own prototype Pcb's I have most of the stuff you need foc . Its just gathering dust in my electronics workshop. We used to design and manufacture interactive audio /video animatronics for Museums

 

[barks944]

I understand the 16V engines are a different version of motronic. But I didn't think the late turbo was any different to the early turbo except for a changes in the size of the EEPROM indicated by a change from 24-28 pin. In fact the later N/A 8V models may also have gone through this change.

1982 to 1985.0 Porsche 944 (ML1.1 & ML1.2)
1985.5 to 1989 Porsche 944/924S (ML3.1) (1988 redesign to 28 pin 8k external EEPROM)
Porsche 944Turbo ML3.1 (Slightly different to the N/A version to work with KLR)
Porsche 944S M2.1
Porsche 944S2 ???
Porsche 968 M2.10.1

 

[barks944]

DME project is coming along nicely, we are finalising the circuit diagrams before we go to PCB layout. It will include USB and bluetooth it should support all the ML3.1 systems including the turbo and N/A models. It will be configurable between the two. Should support wasted spark/boost control/water injection. I have a guy who makes tuning software for saabs to make the tuning software. Adding more stuff to the circuits all the time. Should be able to use a wide array of sensors for mapping including MAP/MAF/TPS/AFM/WBO2/EGT/IAT.

Interested in your PCB prototyping kit! What kind of kit is it?

Can prob convert the PCB layout fairly easily to support the S2 as well, but it will require a different board to the ML3.1.

 

[peanut]

oh nothing special I'm afraid. Don't build your hopes up its not a fully automated pcb maker lol

. I made all the pcb boards we needed (probably over 100) in-house as most were one-off designs .

Basically I drew the circuit with Coreldraw then printed the schematic onto clear acetate. Bought photosensitive pcb board from Maplins or CPC etc and using a ordinary photocopier transferred the acetate onto the pcb board.
Then used an acid bath to dissolve the copper and voila a prototype pcb board. just needed to drill the holes and you had a pcb within a few hours.

I have the flo tubes to make a proper lightbox , electronic timer module and some pcb boards and chemicals etc for diy pcb's which you are welcome to . Later when you have your pcb finalised you will want to get them made professionally but until then these diy boards are as good as the real thing and allow proper testing without mucking about with breadboards and vero board and loads of speedwire etc.

heres a good tutorial.
http://blog.makezine.com/archive/2008/01/how_to_make_printed_circu.html

 

[barks944]

Ahh, we are going surface mount! It's not too bad, I will get 5 or so boards made for testing which brings the price down a bit.

Edit: Actually is there any issue with doing surface mount with that kit? Maybe the accuracy is more of an issue?

 

[peanut]

ORIGINAL: barks944

I understand the 16V engines are a different version of motronic. But I didn't think the late turbo was any different to the early turbo except for a changes in the size of the EEPROM indicated by a change from 24-28 pin. In fact the later N/A 8V models may also have gone through this change.

1982 to 1985.0 Porsche 944 (ML1.1 & ML1.2)
1985.5 to 1989 Porsche 944/924S (ML3.1) (1988 redesign to 28 pin 8k external EEPROM)
Porsche 944Turbo ML3.1 (Slightly different to the N/A version to work with KLR)
Porsche 944S M2.1
Porsche 944S2 ???
Porsche 968 M2.10.1

Porsche 944S2 1988-1990 Bosch Motronic 2.1[]
http://www.ecutesting.com/products/porsche/porsche.html

Although fundamentally they operate very similarly there are a lot of differences between early ECU's and late ECU's as you know. ie ignition amplifier, fuel safety mechanism, cold start mechanism, DME relay operation, , Dwell control, ignition timing
TDC and RPM reference sensors etc. Frustrating to fault find an early car if you are mostly familiar with the later Motronic

 

[peanut]

ORIGINAL: barks944

Ahh, we are going surface mount! It's not too bad, I will get 5 or so boards made for testing which brings the price down a bit.

Edit: Actually is there any issue with doing surface mount with that kit? Maybe the accuracy is more of an issue?

no the only issues are accuracy as you say. We used some SM but its frustrating and time consuming . Space was never an issue for us and using descrete components meant I could use untrained staff for assembly , soldering and testing etc .

good luck with the prototyping. I used to have some excellent programming contacts in Texas and the Far East for our animatronic chip programming but they are probably all gone now.[&o]

 

[peanut]

this is interesting from Wikipedia. about Motronic 1.1 and 1.3 systems


The ECMs have 2 injection outputs, and the injectors are arranged in 2 "banks" which fire once every two engine revolutions. In an example 4-cylinder engine, one output controls the injectors for cylinders 1 and 3, and the other controls 2 and 4.

The system uses a "cylinder ID" sensor mounted to the cam-shaft to detect which cylinders are approaching the top of their stroke, therefore which injector bank should be fired.

During start-up (below 600 RPM), or if there is no signal from the cylinder ID sensor, all injectors are fired simultaneously once per engine revolution.http://en.wikipedia.org/wiki/Motronic#cite_note-ReferenceA-5

this implies that above 600rpm the ECU on 16 valve cars does indeed pulse 1&3 and 2&4 cylinders seperately.?!

Conflicting information from various sources but you have to ask the question why the injector pulsing output from the ECU has been split into two output connections if the injectors do actually fire simultaneously ?

Perhaps it was a option on some ECU models ?