JETRONIC for PIGEON AND REX
Discussion
I was going to respond to the thread on Which Mercedes and CSL, but I thought it would be clearer if I just keep it as a separate entity.
Just to make it clear, the D Jetronic system was the first electronic system from Bosch and, as you stated Pigeon, was very well made. But the single largest problem with it was that there was no facility within the system to read engine air flow; no airflow meter, no air mass meter, no sensor of any type that would tell the ECU how much air the engine was pulling, that is a serious disadvantage in a system.
What you did have was a vacuum sensor which I suppose could be considered a measurement of air input, but of course it can only measure the vacuum as long as it's there; on wide throttle openings and high engine load and speed it becomes redundant.
What you had was a variable throttle sensor connected to the butterfly valve, this gave you a constantly varying reading for the throttle opening; an ambient temperature sensor, this would adjust the mixture slightly for variance in climate changes; this is why Pigeon, you could get away with a resistor in there, because the variance was not great enough that it would drastically affect the running on an engine like a Volvo.
Lastly there was the engine temperature sensors which sometimes was a single unit, sometimes two.
These were the source of the most of the systems problems. Because they altered the mixture so widely, they had to be functioning perfectly and, by the same token, so had the cooling system to keep the temperature within the correct limits that the ECU was programmed to.
When I had my last 3.0 CSi, I took the head off to gas flow it and popped all the core plugs from the block, cleaned the cooling channels until they were shining, fitted a new water pump, radiator, heater matrix and thermostat, all genuine BMW.
I also got rid of the thermo coupled cooling fan, which was a source of overheating on these early cars, and fitted an electric fan from a 735i.
After I built it I renewed the temperature sensor and used the correct Bosch diagnostic equipment to set everything up; it was utterly perfect after that, faultless to drive, but of course theoretically it was better than new!!
Now we come to the K Jetronic. Pigeon, I am really surprised that a mechanically sympathetic guy like yourself doesn't endorse the K Jetronic system, but I suspect it's simply because you haven't encountered it properly and aren't aware of just how good it is.
You have to be careful when you call it the mechanical system, because it can be confused with the old mechanical direct injection that Rex talked about on his Merc W114. K Jetronic is totally different and is quite simply the most brilliant fuel system ever devised.
The references that Rex made to high fuel consumption I'm afraid are totally inaccurate, and must be due to maladjustment or some other fault.
The two best examples to illustrate this are the very first low Cd Audi 100 2.2 saloon and the very early BMW 520 E28 Saloon.
Both of these cars were introduced in '81 and were milestones in what they achieved and represented in terms of technical development.
The Audi was the lowest drag car ever manufactured at that time and would return almost 50 MPG at 56 MPH, this is 1981 remember!!
It was running on K Jetronic with only a small electronic control relay for finite ignition and idle control.
The BMW was developed using the 2 litre lightweight belt driven overhead cam engine that had been in the E12 5 series but with the same K Jetronic system that the Audi had. BMW had already used this in the 323i with tremendous success, but for unknown reasons only used it in the 5 series for a couple of years before adopting Motronic.
The 520 with K Jetronic was far more reliable, smoother, nicer to drive and surprisingly more economical than the Motronic version!
K Jetronic works by utilising a large butterfly valve attached by a pivot to an extremely accurately made, (tenth of a thou tolerances!) fuel metering unit situated in the airstream to the inlet manifold. As the engine draws more air the valve is lifted and consequentially moves a plunger in the fuel unit which uncovers pressurised slots or chambers within the unit allowing fuel to flow to the injectors; the advantage being that the fuel flow to all injectors is identical and is in direct proportion to the engine's needs because it's controlled by the airflow.
There are various sensors and a pressure regulator to give more finite control for different conditions such as cold start, but the beauty of the system is that if you have a supply to the fuel pump the engine will run, even if the electronics break down or give problems, as long as you have fuel pressure the system will fire up the engine.
The system was finally developed to become KE Jetronic, incorporating extra control valves for the fuel metering unit in conjunction with a Lamda sensor and a cat, and that version gave economy and refinement that was truly outstanding.
Now Rex, you should have no problem whatsoever fitting K Jetronic to your 323i coz that's what the original engine had anyway, you should still be able to pick up all the components required.
The trick when installing K Jetronic is the intake cone shape of the butterfly housing; this determines the variance of the fuel flow at different engine conditions and states of tune.
The depth, angles and number of facets to the cone determines how much fuel the engine gets and when it gets it!
You shold still be able to purchase some superb books on tuning Bosch injection systems on Amazon; in these you will find the data and formulas necessary to get the cones right.
I would love to fit a K Jetronic system to a 3 litre coupe, it would be soooo smooth and tractable!
I'm gonna have to make a coffee after that lot!!
Just to make it clear, the D Jetronic system was the first electronic system from Bosch and, as you stated Pigeon, was very well made. But the single largest problem with it was that there was no facility within the system to read engine air flow; no airflow meter, no air mass meter, no sensor of any type that would tell the ECU how much air the engine was pulling, that is a serious disadvantage in a system.
What you did have was a vacuum sensor which I suppose could be considered a measurement of air input, but of course it can only measure the vacuum as long as it's there; on wide throttle openings and high engine load and speed it becomes redundant.
What you had was a variable throttle sensor connected to the butterfly valve, this gave you a constantly varying reading for the throttle opening; an ambient temperature sensor, this would adjust the mixture slightly for variance in climate changes; this is why Pigeon, you could get away with a resistor in there, because the variance was not great enough that it would drastically affect the running on an engine like a Volvo.
Lastly there was the engine temperature sensors which sometimes was a single unit, sometimes two.
These were the source of the most of the systems problems. Because they altered the mixture so widely, they had to be functioning perfectly and, by the same token, so had the cooling system to keep the temperature within the correct limits that the ECU was programmed to.
When I had my last 3.0 CSi, I took the head off to gas flow it and popped all the core plugs from the block, cleaned the cooling channels until they were shining, fitted a new water pump, radiator, heater matrix and thermostat, all genuine BMW.
I also got rid of the thermo coupled cooling fan, which was a source of overheating on these early cars, and fitted an electric fan from a 735i.
After I built it I renewed the temperature sensor and used the correct Bosch diagnostic equipment to set everything up; it was utterly perfect after that, faultless to drive, but of course theoretically it was better than new!!
Now we come to the K Jetronic. Pigeon, I am really surprised that a mechanically sympathetic guy like yourself doesn't endorse the K Jetronic system, but I suspect it's simply because you haven't encountered it properly and aren't aware of just how good it is.
You have to be careful when you call it the mechanical system, because it can be confused with the old mechanical direct injection that Rex talked about on his Merc W114. K Jetronic is totally different and is quite simply the most brilliant fuel system ever devised.
The references that Rex made to high fuel consumption I'm afraid are totally inaccurate, and must be due to maladjustment or some other fault.
The two best examples to illustrate this are the very first low Cd Audi 100 2.2 saloon and the very early BMW 520 E28 Saloon.
Both of these cars were introduced in '81 and were milestones in what they achieved and represented in terms of technical development.
The Audi was the lowest drag car ever manufactured at that time and would return almost 50 MPG at 56 MPH, this is 1981 remember!!
It was running on K Jetronic with only a small electronic control relay for finite ignition and idle control.
The BMW was developed using the 2 litre lightweight belt driven overhead cam engine that had been in the E12 5 series but with the same K Jetronic system that the Audi had. BMW had already used this in the 323i with tremendous success, but for unknown reasons only used it in the 5 series for a couple of years before adopting Motronic.
The 520 with K Jetronic was far more reliable, smoother, nicer to drive and surprisingly more economical than the Motronic version!
K Jetronic works by utilising a large butterfly valve attached by a pivot to an extremely accurately made, (tenth of a thou tolerances!) fuel metering unit situated in the airstream to the inlet manifold. As the engine draws more air the valve is lifted and consequentially moves a plunger in the fuel unit which uncovers pressurised slots or chambers within the unit allowing fuel to flow to the injectors; the advantage being that the fuel flow to all injectors is identical and is in direct proportion to the engine's needs because it's controlled by the airflow.
There are various sensors and a pressure regulator to give more finite control for different conditions such as cold start, but the beauty of the system is that if you have a supply to the fuel pump the engine will run, even if the electronics break down or give problems, as long as you have fuel pressure the system will fire up the engine.
The system was finally developed to become KE Jetronic, incorporating extra control valves for the fuel metering unit in conjunction with a Lamda sensor and a cat, and that version gave economy and refinement that was truly outstanding.
Now Rex, you should have no problem whatsoever fitting K Jetronic to your 323i coz that's what the original engine had anyway, you should still be able to pick up all the components required.
The trick when installing K Jetronic is the intake cone shape of the butterfly housing; this determines the variance of the fuel flow at different engine conditions and states of tune.
The depth, angles and number of facets to the cone determines how much fuel the engine gets and when it gets it!
You shold still be able to purchase some superb books on tuning Bosch injection systems on Amazon; in these you will find the data and formulas necessary to get the cones right.
I would love to fit a K Jetronic system to a 3 litre coupe, it would be soooo smooth and tractable!
I'm gonna have to make a coffee after that lot!!
You are right that I haven't really encountered the K-Jetronic properly; never had to fix one, let alone owned one. I think my feeling arises from the fact that I prefer my analogue computers to be electronic rather than mechanical, again because that's the form with which I am most familiar - I'd be much happier implementing a reasonably complex mathematical function with a bunch of op-amps and p-n junctions, and designing around potential causes of inaccuracy or drift, than I would be doing it mechanically; the same goes for the various sensors and actuators. (I do regret never having taken my D-Jetronic ECU apart to see what Bosch had done... (fx: hits google) Ah, but this guy has: http://members.rennlist.com/pbanders/ looks like I've got some reading to do!) Then there's the matter of predicting the performance of a design; the equations of electronic circuitry are a lot more friendly than those of fluid dynamics 
Pigeon, I must be really sleeping tonight. I just realised that what Rex is talking about with his BMW is to fit the original old mechanical injection system in PLACE OF K Jetronic.
I can't think of why he would want to do this, but it's a helluva task to get it right.
The mechanical system has a timed injection pump that's driven from the crankshaft just like a diesel, and the mixture is controlled by a combination of a pushrod from the throttle linkage, a waxstat temperature sensor and a vacuum diaphragm.
BMW used it on the 2002 Tii to fantastic effect, it was seriously rapid, and Mercedes also used it on lots of models.
He would have to find a mechanical injection pump for a six cylinder engine and have it calibrated to suit the 323 engine; then mount it to the block and have a drive pulley made for the crankshaft.
Then there's the question of the injection timing which would have to be calculated, and even then a certain amount of experimentation would be essential.
Then he'd have to get the injectors into the head somehow.
Still, it would be quite a project to have a crack at!!
Just a wee note about the D Jetronic ECUs Pigeon.
These were not micro chip of course at that age, the beauty of that being that they were repairable on the bench, but they did tend to be a bit delicate and very susceptible to dampness problems.
To get round this some manufacturers, Jaguar being one, filled the ECU after manufacture with resin wax to totally seal it, which probably did help to keep dampness problems away, but rendered the damn thing useless afterwards!!
What is a lot better is the upgraded system on the later cars which was L Jetronic; this was D Jetronic with an airflow meter added in and a suitably modified ECU.
It was hugely successful and fitted to dozens of cars, virtually the whole BMW range, Mantas, Monzas, Fords, etc. It can be very nicely tuned by modifying the airflow meter and the ECU, so you could probably enjoy monkeying about with that.
In fact you could quite easily upgrade your Volvo onto L Jetronic and further enhance the economy.
I can't think of why he would want to do this, but it's a helluva task to get it right.
The mechanical system has a timed injection pump that's driven from the crankshaft just like a diesel, and the mixture is controlled by a combination of a pushrod from the throttle linkage, a waxstat temperature sensor and a vacuum diaphragm.
BMW used it on the 2002 Tii to fantastic effect, it was seriously rapid, and Mercedes also used it on lots of models.
He would have to find a mechanical injection pump for a six cylinder engine and have it calibrated to suit the 323 engine; then mount it to the block and have a drive pulley made for the crankshaft.
Then there's the question of the injection timing which would have to be calculated, and even then a certain amount of experimentation would be essential.
Then he'd have to get the injectors into the head somehow.
Still, it would be quite a project to have a crack at!!
Just a wee note about the D Jetronic ECUs Pigeon.
These were not micro chip of course at that age, the beauty of that being that they were repairable on the bench, but they did tend to be a bit delicate and very susceptible to dampness problems.
To get round this some manufacturers, Jaguar being one, filled the ECU after manufacture with resin wax to totally seal it, which probably did help to keep dampness problems away, but rendered the damn thing useless afterwards!!
What is a lot better is the upgraded system on the later cars which was L Jetronic; this was D Jetronic with an airflow meter added in and a suitably modified ECU.
It was hugely successful and fitted to dozens of cars, virtually the whole BMW range, Mantas, Monzas, Fords, etc. It can be very nicely tuned by modifying the airflow meter and the ECU, so you could probably enjoy monkeying about with that.
In fact you could quite easily upgrade your Volvo onto L Jetronic and further enhance the economy.
IOLAIRE, I agree with you and LOVE the K Jet system.
I know it very well and already have it fitted to my 2.7 litre E21 BMW. WIth a larger ignition vacuum capsule and a distibutor with more low speed advance and less top end- (I've "bumped" the curve by mucking about with the springs)- I actually get around 33 mpg on a spirited run on the motorway! I MUCH MUCH prefer K Jet to the L Jet system. I imagine BMW went to L jet either for emissions reasons or due to some corporate Strategy move- at the time they always wanted to be seen sporting the latest technologies (result= early motornic 7 series was the MOST seen on the hard shoulder in the early eighties!)
My biggest problem is that I want to fit port throtles onto my 3.1 litre M20 complete with the 14 litre plenum of the E36 M3 engine for good top end. With the metering vein of the air mass meter of K Jetronic fitted to this- the transient fueling response will lag so badly behind the throttled air response I will get lean misfire on transient accels and rich spots on tip-outs.
The vacuum line that fits into the Warm up regulator- would have to be mounted near one of the port throttles- and the problem with this approach ( I leanred from an alternative fuel project I was working on)- is that the pressure signals this near to the runner/inlet valve will be wildly fluctuating. Even if you Tee off 6 pipes - this is unlikely to solve the problem.
The only hope I have are two fold:
Porsche were FORCED to retain K Jetronic FI on their racing 934s complete with port throttles- so I'm trying to resserch how hey got around the problem.
The other hope is that mid 70's US market K jetronic 911 Porsches did NOT use the vacuum line in the WUR to richen up the mixture but a rod type/cam actuator from the throttle(s). If I can find out more about this system and how it worked on transients- it might be my salvation- in keeping my favoured K Jet system.
I'm not worried about the diameter of the cone/fuel supply of the cast iron fuel distributor. I know a company that can rebuild them to my specs, and I've come to terms with the maths of the flow of these. I have a CFD package I use as a "steady state rig" to do the rough calcs- while the fine tuning can be done by the place that would rebuild the system. So I should be able to come up with an amalgam of using a Porsche 4.5 litre V8 928 K jet air cone mated to a 6 cylinder fuel distributor (Probably off of a Porsche 930 Turbo- but with the idle/part load/full load region remachined according to my calcs).
I know what kind of top end volumetrc efficiencies I'm looking for and from this I can work out my air flow and therefore my required fuel flow.
I was considering the purely Bosch plunger pump mechanical system- as used on the early 911 Porsches- as a contingency- as this IS compatible with port throttles.
PS: Anyone telling me to fit some after market mapable modern Injection system such as Megasquirt or MBE will be ignored- I know how to map modern systems and don't need to be patronised be being told of the benefits- and am not interested in this for THIS project!
I know it very well and already have it fitted to my 2.7 litre E21 BMW. WIth a larger ignition vacuum capsule and a distibutor with more low speed advance and less top end- (I've "bumped" the curve by mucking about with the springs)- I actually get around 33 mpg on a spirited run on the motorway! I MUCH MUCH prefer K Jet to the L Jet system. I imagine BMW went to L jet either for emissions reasons or due to some corporate Strategy move- at the time they always wanted to be seen sporting the latest technologies (result= early motornic 7 series was the MOST seen on the hard shoulder in the early eighties!)
My biggest problem is that I want to fit port throtles onto my 3.1 litre M20 complete with the 14 litre plenum of the E36 M3 engine for good top end. With the metering vein of the air mass meter of K Jetronic fitted to this- the transient fueling response will lag so badly behind the throttled air response I will get lean misfire on transient accels and rich spots on tip-outs.
The vacuum line that fits into the Warm up regulator- would have to be mounted near one of the port throttles- and the problem with this approach ( I leanred from an alternative fuel project I was working on)- is that the pressure signals this near to the runner/inlet valve will be wildly fluctuating. Even if you Tee off 6 pipes - this is unlikely to solve the problem.
The only hope I have are two fold:
Porsche were FORCED to retain K Jetronic FI on their racing 934s complete with port throttles- so I'm trying to resserch how hey got around the problem.
The other hope is that mid 70's US market K jetronic 911 Porsches did NOT use the vacuum line in the WUR to richen up the mixture but a rod type/cam actuator from the throttle(s). If I can find out more about this system and how it worked on transients- it might be my salvation- in keeping my favoured K Jet system.
I'm not worried about the diameter of the cone/fuel supply of the cast iron fuel distributor. I know a company that can rebuild them to my specs, and I've come to terms with the maths of the flow of these. I have a CFD package I use as a "steady state rig" to do the rough calcs- while the fine tuning can be done by the place that would rebuild the system. So I should be able to come up with an amalgam of using a Porsche 4.5 litre V8 928 K jet air cone mated to a 6 cylinder fuel distributor (Probably off of a Porsche 930 Turbo- but with the idle/part load/full load region remachined according to my calcs).
I know what kind of top end volumetrc efficiencies I'm looking for and from this I can work out my air flow and therefore my required fuel flow.
I was considering the purely Bosch plunger pump mechanical system- as used on the early 911 Porsches- as a contingency- as this IS compatible with port throttles.
PS: Anyone telling me to fit some after market mapable modern Injection system such as Megasquirt or MBE will be ignored- I know how to map modern systems and don't need to be patronised be being told of the benefits- and am not interested in this for THIS project!
IOLAIRE said:
I just realised that what Rex is talking about with his BMW is to fit the original old mechanical injection system in PLACE OF K Jetronic.
I can't think of why he would want to do this, but it's a helluva task to get it right.
The mechanical system has a timed injection pump that's driven from the crankshaft just like a diesel, and the mixture is controlled by a combination of a pushrod from the throttle linkage, a waxstat temperature sensor and a vacuum diaphragm.
BMW used it on the 2002 Tii to fantastic effect, it was seriously rapid, and Mercedes also used it on lots of models.
He would have to find a mechanical injection pump for a six cylinder engine and have it calibrated to suit the 323 engine; then mount it to the block and have a drive pulley made for the crankshaft.
Then there's the question of the injection timing which would have to be calculated, and even then a certain amount of experimentation would be essential.
Then he'd have to get the injectors into the head somehow.
I'm looking into the the pulley problem right now- wondering whether I can drive it off the intermediate shaft drive. It could be a problem. Fitting the injectors won't be a problem-as I'm using the E36 M3 inlet manfiold, but fabricating up extra inlet runner flanges- I'll either have them made via coated stereolithography or via aluminium. Injection timing won't be a problem either- I've completed several injection timing sensitivities on Jag Engines- and if you're not concerned about emissions performance- it's of little consequence.
It's true you'll lose some bottom end charge cooling- hence volumetric efficiency- but this will be slight. At top end of course. even timed injection system pretty much run close to continuously anyway. I intend to have AFR meters installed at the exhaust manifold/header junctions to make sure my fueling is roughly on par.
IOLAIRE, I think I was half asleep too as I made pretty much the same oversight...
The dieselesque system - would I be right in thinking that this was adapted, or at least branded, by Lucas and became the system Triumph used in the six-cylinder TRs and the 2.5PI? A significant problem being wear in the metering unit, which I would have thought would be expected, given that the early experiments with petrol injection were based on diesel technology and suffered the same problem, due to the lack of lubricating qualities of petrol? Or am I getting totally mixed up?
I've been reading through the D-Jet info at the URL I posted earlier, and it's fascinating stuff. I didn't realise that the manifold air pressure sensor was not just a sensor, but actually performs some of the computation by means of differing spring rates. There is also quite a bit of detail on the analogue methods used to synthesise the volumetric efficiency curve. Full circuit diagrams of the ECU as well, which will provide some interesting bedtime reading
The idea of upgrading a Volvo installation to L-Jet does indeed sound fun. I no longer have the Amazon I've been talking about - this was quite a long time ago - but as I've mentioned, the idea of getting a 164E is looking quite attractive once my automotive projects fund has got a little larger.
The dieselesque system - would I be right in thinking that this was adapted, or at least branded, by Lucas and became the system Triumph used in the six-cylinder TRs and the 2.5PI? A significant problem being wear in the metering unit, which I would have thought would be expected, given that the early experiments with petrol injection were based on diesel technology and suffered the same problem, due to the lack of lubricating qualities of petrol? Or am I getting totally mixed up?
I've been reading through the D-Jet info at the URL I posted earlier, and it's fascinating stuff. I didn't realise that the manifold air pressure sensor was not just a sensor, but actually performs some of the computation by means of differing spring rates. There is also quite a bit of detail on the analogue methods used to synthesise the volumetric efficiency curve. Full circuit diagrams of the ECU as well, which will provide some interesting bedtime reading
The idea of upgrading a Volvo installation to L-Jet does indeed sound fun. I no longer have the Amazon I've been talking about - this was quite a long time ago - but as I've mentioned, the idea of getting a 164E is looking quite attractive once my automotive projects fund has got a little larger.
Marquis_Rex said:
IOLAIRE, I agree with you and LOVE the K Jet system.
I know it very well and already have it fitted to my 2.7 litre E21 BMW. WIth a larger ignition vacuum capsule and a distibutor with more low speed advance and less top end- (I've "bumped" the curve by mucking about with the springs)- I actually get around 33 mpg on a spirited run on the motorway! I MUCH MUCH prefer K Jet to the L Jet system. I imagine BMW went to L jet either for emissions reasons or due to some corporate Strategy move- at the time they always wanted to be seen sporting the latest technologies (result= early motornic 7 series was the MOST seen on the hard shoulder in the early eighties!)
My biggest problem is that I want to fit port throtles onto my 3.1 litre M20 complete with the 14 litre plenum of the E36 M3 engine for good top end. With the metering vein of the air mass meter of K Jetronic fitted to this- the transient fueling response will lag so badly behind the throttled air response I will get lean misfire on transient accels and rich spots on tip-outs.
The vacuum line that fits into the Warm up regulator- would have to be mounted near one of the port throttles- and the problem with this approach ( I leanred from an alternative fuel project I was working on)- is that the pressure signals this near to the runner/inlet valve will be wildly fluctuating. Even if you Tee off 6 pipes - this is unlikely to solve the problem.
The only hope I have are two fold:
Porsche were FORCED to retain K Jetronic FI on their racing 934s complete with port throttles- so I'm trying to resserch how hey got around the problem.
The other hope is that mid 70's US market K jetronic 911 Porsches did NOT use the vacuum line in the WUR to richen up the mixture but a rod type/cam actuator from the throttle(s). If I can find out more about this system and how it worked on transients- it might be my salvation- in keeping my favoured K Jet system.
I'm not worried about the diameter of the cone/fuel supply of the cast iron fuel distributor. I know a company that can rebuild them to my specs, and I've come to terms with the maths of the flow of these. I have a CFD package I use as a "steady state rig" to do the rough calcs- while the fine tuning can be done by the place that would rebuild the system. So I should be able to come up with an amalgam of using a Porsche 4.5 litre V8 928 K jet air cone mated to a 6 cylinder fuel distributor (Probably off of a Porsche 930 Turbo- but with the idle/part load/full load region remachined according to my calcs).
I know what kind of top end volumetrc efficiencies I'm looking for and from this I can work out my air flow and therefore my required fuel flow.
I was considering the purely Bosch plunger pump mechanical system- as used on the early 911 Porsches- as a contingency- as this IS compatible with port throttles.
PS: Anyone telling me to fit some after market mapable modern Injection system such as Megasquirt or MBE will be ignored- I know how to map modern systems and don't need to be patronised be being told of the benefits- and am not interested in this for THIS project!
Hi Rex,
I think I have come up with a possible solution to running K Jet with port throttles.
You're right, Porsche did do this with the 911s and I have also definitely seen a racing CSL with a similar set up.
I'll try and post back on here tomorrow and explain it, very busy tonight.
Thanks IOLAIRE-any help would be appreciated,but no hurry.
I would rather stay with K jet if I could- it is my preffered FI system. So many systems around to mix and match with ( provided one knows what they're doing), so much background. Contrary to popular belief, also, the loss across the generously dimensioned Air mass Funnel isn't that great compared to L jetronic. It is obviously higher then something like the Bosch mechanical/Kugel Fischer system which doesn't need any ducting to a air flow measuring device at all- but by the same token they need frequent re-calibrating!
I would rather stay with K jet if I could- it is my preffered FI system. So many systems around to mix and match with ( provided one knows what they're doing), so much background. Contrary to popular belief, also, the loss across the generously dimensioned Air mass Funnel isn't that great compared to L jetronic. It is obviously higher then something like the Bosch mechanical/Kugel Fischer system which doesn't need any ducting to a air flow measuring device at all- but by the same token they need frequent re-calibrating!
Excellent reading, I'm enjoying this new vintage forum .
As for the K-Jetronic, I have it in my '86 Mercedes. It has never been a problem and makes a smooth revving engine (if a little noisy without balancer shafts!) with a torquey, linear power delivery. It's interesting that it can run so long as you have fuel pressure.
Can you tell us some more about the ancillaries - how does cold start work? I know there's a pre-start injector for when the engine's cold, and an engine temperature sensor. My car has always been missing any sort of cold start feature - I'm sure it should be there, but just doesn't work on my car, so at idle it stumbles and eventually stalls.
What system do the '90s VWs use?
regards
Russ
'86 190E K-jetronic, '62 Elan Carburettors
. As for the K-Jetronic, I have it in my '86 Mercedes. It has never been a problem and makes a smooth revving engine (if a little noisy without balancer shafts!) with a torquey, linear power delivery. It's interesting that it can run so long as you have fuel pressure.
Can you tell us some more about the ancillaries - how does cold start work? I know there's a pre-start injector for when the engine's cold, and an engine temperature sensor. My car has always been missing any sort of cold start feature - I'm sure it should be there, but just doesn't work on my car, so at idle it stumbles and eventually stalls.
What system do the '90s VWs use?
regards
Russ
'86 190E K-jetronic, '62 Elan Carburettors
The problems that I had with K-Jet on my 8 valve Saab Turbo's (spent literally weeks troubleshooting on occasion, for instance where an engine would flood on start up) were indicative of a 'mechanical' system after 15 or so years from new, however none were related to the actual K-Jet fuel distributor which is indeed a fine piece of mechanical engineering. I'm afraid though, the same can't be said about the 'warm up regulator' (control pressure regulator that doubles as a boost pressure enrichment device) or the thermo time swtich which controls the cold start injector...
IOLAIRE said:
But the single largest problem with it was that there was no facility within the system to read engine air flow; no airflow meter, no air mass meter, no sensor of any type that would tell the ECU how much air the engine was pulling, that is a serious disadvantage in a system.
What you did have was a vacuum sensor which I suppose could be considered a measurement of air input, but of course it can only measure the vacuum as long as it's there; on wide throttle openings and high engine load and speed it becomes redundant.
A great number of modern EFI systems use the MAP sensor which you have described. The output of the sensor is used in conjunction with engine RPM and a volumetric efficiency map within the ECU to determine the amount of air the engine is using. If calibrated properly this system is just as accurate as an AFM, and considerably more robust. In fact from memory, I think it actually has a greater dymanic range than most AFM (which are not normaly usable under very low airflow conditions).
>> Edited by MR2Mike on Thursday 30th December 00:45
That's exactly what D-Jet does. The thing is that the accuracy of the mapping is affected by a great many factors. It seems that I was lucky with my system in that both the injection system itself and the engine were in pretty good condition so everything was working as designed, but there is no means of compensating for wear, drift or any other deficiency - it's quite hard to think of anything that can change over time in the engine or its major ancillaries that doesn't affect volumetric efficiency to some extent. And of course if you want to do any tuning work you're buggered, whereas an AFM would work fine (as long as you didn't go too wild and out of its range).
Off the top of my head I can think of two reasons why a modern system could be more successful here: in the short term, the MAP sensor / VE map can provide a "close-enough" point from which the mixture can be adjusted using feedback from the lambda sensor; and in the long term it can "learn" from this same feedback and tweak the map to start from a closer approximation next time. Neither advantage was enjoyed by the entirely analogue, pre-lambda-sensor D-Jet.
Off the top of my head I can think of two reasons why a modern system could be more successful here: in the short term, the MAP sensor / VE map can provide a "close-enough" point from which the mixture can be adjusted using feedback from the lambda sensor; and in the long term it can "learn" from this same feedback and tweak the map to start from a closer approximation next time. Neither advantage was enjoyed by the entirely analogue, pre-lambda-sensor D-Jet.
Gassing Station | Classic Cars and Yesterday's Heroes | Top of Page | What's New | My Stuff