MOTOBRICK.COM
TECHNICAL MOTOBRICK WRENCHING In Remembrance of Inge K. => The Motobrick Workshop => Topic started by: Mongrel on December 29, 2016, 01:44:38 PM
-
I've been searching for a while, trying to satisfy my curiosity, without finding an answer - has anyone yet built a custom replacement for the Jetronic computer? Say, in a café build where they're shaving every bit of space off the bike they can? I feel like I've seen that in some build report somewhere, but can't say for sure.
-
There is one -- the BSK Speedworks Engine Management Unit (http://www.bskspeedworks.co.uk/engine-management-package.html). However, it doesn't come cheap. There have been persons who have managed to adapt the Megasquirt ECU to the K75 which will replace part of the Jetronic system (the ICU but not the ECU). The Jetronic is an open loop analog system. Consequently, there are a number of deficiencies in the motor that preclude adapting a more modern digital engine management unit to the bike. A more modern unit would need more sensors to feedback engine status and those just don't exist on a stock motor.
-
Rob, it's an interesting question. Another question that comes to mind is what would be involved in replacing a Jetronic with a Motronic unit? As far as I can tell, the throttle position sensors need to be swapped, as well as the temperature sensor.
You would need to add a Lambda sensor to the exhaust, and you can remove the air flow sensor. Obviously, a bunch of rewiring of the engine harness will be necessary.
What else might need to be done? Seeing the tight throttle response and better fuel economy of the Motronic, I wouldn't mind swapping the systems on my K75's.
-
you could go this route.....
-
Rob, it's an interesting question. Another question that comes to mind is what would be involved in replacing a Jetronic with a Motronic unit? As far as I can tell, the throttle position sensors need to be swapped, as well as the temperature sensor.
You would need to add a Lambda sensor to the exhaust, and you can remove the air flow sensor. Obviously, a bunch of rewiring of the engine harness will be necessary.
What else might need to be done? Seeing the tight throttle response and better fuel economy of the Motronic, I wouldn't mind swapping the systems on my K75's.
of course, the motronic expects that fourth cylinder which a 75 doesn't have.
early versions of motronic used the same design air flow meter that that jetronic does. can't get rid of it.
would also need to an an O2 sensor (narrow band) to the exhaust system.
you could delete the AFM if you ran an Alpha-N type fuel management system as a piggy back to the jetronic. that adds another ECU and a wideband O2 to the mix. the AN computer runs the fueling, the jetronic(or motronic) runs the timing. that is a similar setup to what i have on the S14 engine in my track car.
-
Thanks Rob.
We got curious because a number of my friends were talking antique restorations and one particular curmudgeon (who never does rebuilds on older cars, only tinkering with new ones) declared that antique restoration would never work for post-1980 cars with all their electronics and computers with highly specific firmware "lost to the ages", the idea being that if you need, say, an ABS control unit for your '94 Camry rebuild in 2090 (assuming we didn't blow the planet up or outlaw all cars or anything) you'll just be stuck.
Now a number of us already pointed out all the various reasons that was silly, especially that fabricating some sort of replacement is already possible now, let alone with 75+ more years of technology. But it got me curious because a K needing a replacement for the Jetronic was of course the first example which came to mind and I couldn't actually recall if that had already happened in the case of the K, though I felt like I'd seen such a thing somewhere (but then, memory is hardly trustworthy... especially mine).
-
"of course, the motronic expects that fourth cylinder which a 75 doesn't have.
early versions of motronic used the same design air flow meter that that jetronic does. can't get rid of it.
would also need to an an O2 sensor (narrow band) to the exhaust system."
Does it really need that 4th cylinder? With an injector on each cylinder and fuel delivery controlled by pulse width modulation, it is logical that the number of cylinders is not important to the operation of the system.
Also, with it's throttle position sensor, airflow is inferred by throttle butterfly position. Again, since the system no longer considers total air mass via the MAF which supplies all cylinders, but rather the amount that is going into the individual cylinder, the Motronic doesn't care how many cylinders there are.
Looking at my bikes, the only external hardware difference between the Motronic and Jetronic bikes is in the measurement of airflow and exhaust gas. They both have Hall Effect Sensors, injectors and a water temperature sensor. From my experience with my bikes, it appears that the throttle position switch of the Jetronic is a simple mechanical swap to the Motronic's throttle position sensor.
Adding an O2 sensor is a matter of making a hole and welding a bung into the muffler. Since the exhaust does not have a catalytic converter, it may be more desirable to go with the Motronic that uses an idle mixture pot like my K100RS has. Still, there may be a way to use an O2 sensor to close the loop even in the absence of a catalytic converter.
Like the throttle position switch/sensor, the temperature sensor should also be a simple part swap.
The large question that I am not sure how to approach is the ignition coils. With the K75 having a coil per plug, and the K100 having a coil for two plugs, timing could be a problem, not to mention matching coil loads to the Motronic's ignition amplifier. The K100 has a wasted spark ignition which could also work on the K75.
Another question related to ignition is how the Hall Effect Sensors time the system's ignition output both in the Motronic and the Jetronic systems. If there is a point where the two systems can't be made compatible, I suspect this is it.
While I am reasonably familiar with the external components, it's the inside of the Motronic that I am severely deficient on.
-
Megasquirt/MicroSquirt are capable of handling fueling and ignition on 3 or 4 cylinder K's...
There is a thread in one of their forums about ignition on a turbo K...
-
you could go this route.....
Can you provide a link to the source of that photo, Chaos, or is that some of your handiwork? :giggles
-
Can you provide a link to the source of that photo, Chaos, or is that some of your handiwork? :giggles
Fellow across the pond who had a turbo on his K, then decided to simplify his life with a single bing http://www.ukgser.com/forums/showthread.php/428758-BMW-K100-with-single-carburettor
-
Fellow across the pond who had a turbo on his K, then decided to simplify his life with a single bing
Thanks, Chaos! Altering the bores to balance cylinders 1 and 4 must have take patience—and maybe more than one manifold version.
-
Thanks, Chaos! Altering the bores to balance cylinders 1 and 4 must have take patience—and maybe more than one manifold version.
He probably put more effort into that than the limey engineers whom designed my Spitfire Mk IV's manifold :hehehe
-
He probably put more effort into that than the limey engineers whom designed my Spitfire Mk IV's manifold :hehehe
After a yellow fan has been installed, further effort is wasted.
-
I've done a microsquirt on my K75. Did about 5000 miles on it over a year and a half with the microsquirt. Ran it with ITB Mode, mixing alpha-N at high load, and Speed-Density at low load. In short, it worked. Quite well overall. Had better low end fuelling, and perhaps better fuel mileage. But never as trustworthy as the Jetronic, and my perfectionism constantly made me want to fidget with the fuelling. So I went back to stock.
-
Chaos,thanks for posting those links.Yeah,I could go single Bing!
I have it on good authority the Limey engineers would drink heavily before and whilst they designed these manifolding afterthoughts.
-
Chaos,thanks for posting those links.Yeah,I could go single Bing!
I have it on good authority the Limey engineers would drink heavily before and whilst they designed these manifolding afterthoughts.
Possibly a better option than the Bing is to go with the Keihin off a modern Harley. Very solid, accurate, and non-finicky CV carb. Very cheap too: people take them off their bikes to put screaming beagle replacements on.
-
Possibly a better option than the Bing is to go with the Keihin off a modern Harley. Very solid, accurate, and non-finicky CV carb. Very cheap too: people take them off their bikes to put screaming beagle replacements on.
Noticed this comment on the Limey thread:
"After having to make three manifolds, you'll probably understand why I don't want to give out exact dimensions!
All I'll say is that you'll have to alter the inside diameter of each 'runner' to equalise flow. "
He claims "the bike is running really well" (translation into American: "Runs good."), but it must have taken some doing to get to that point.
Hey WMAX, nice to see you.
-
I've been pretty busy with med school, but I try to pop in and share stuff on the megasquirt and what not.
Given the complexities of the flow in a manifold (with a sharp turn as well, assuming you don't use a downdraft carb), I could see it being easier to just stick with EFI and doing a megasquirt.
Could be fun to do some computation modeling and make a 3d-printed equal length runner manifold. Maybe impractical, but still would be cool.
-
That Bing equipped K made me remember that Luftmeister once sold Keihin (?) carbs for the K and Fuel Injection for Airheads... :mbird
-
I've done a microsquirt on my K75. Did about 5000 miles on it over a year and a half with the microsquirt. Ran it with ITB Mode, mixing alpha-N at high load, and Speed-Density at low load. In short, it worked. Quite well overall. Had better low end fuelling, and perhaps better fuel mileage. But never as trustworthy as the Jetronic, and my perfectionism constantly made me want to fidget with the fuelling. So I went back to stock.
Shame you got that far and stopped. I noted your post in the K mods section and wondered what had developed in conclusion.
I have over 5 years of successful operation on a 83 R100RT that includes forced induction and lots of other enhancements on MicroSquirt.
As a new member here, I've started a post to document some mods on a 95 K75, and will include a MicroSquirt and forced induction (in due course, sort of a slow project, stress reducer)
-
Shame you got that far and stopped. I noted your post in the K mods section and wondered what had developed in conclusion.
I have over 5 years of successful operation on a 83 R100RT that includes forced induction and lots of other enhancements on MicroSquirt.
As a new member here, I've started a post to document some mods on a 95 K75, and will include a MicroSquirt and forced induction (in due course, sort of a slow project, stress reducer)
Yeah, for me, and my bone-stock K75, it just didn't add enough to be worth the extra work. The reliability, simplicity, and known water-resistance of the Jetronic was more important for me. Especially because I started to commute a lot on it, and now commute on it rain or shine, cold or hot, without a place to park a car at school if it doesn't run.
Look forward to reading what you do with the K75. Let me know if you have any questions on how I did things.
-
Well, I will most certainly take you up on your offer as things get going.
Out of curiousity, can you further explain your concern with "water resistance" on the K mod you had developing? I have as mentioned, a long running MicroSquirt system commissioned on my 83 R100RT, and the ignition components (crank & cam chopper wheels and VR sensors etc. etc.) are all located at the very front of the engine timing chest and in direct line of fire to road spray and frontal rain blasting. I have been through some significant deluges and horrendous West Coast rain storms - never ever had a problem. In fact been on long distance trips about the continent with nary an issue (touch wood).
Perhaps something else you unearthed for the K, but my experience is better reliability than stock (had a trip spoil after my brother lost the Hall in his airhead bean can in Oregon and we could not seem to locate parts & trailered her home)
-
There are a number of options listed on the German flyingbrick.de site. If you're not registered you may not have access, but check out here:
http://www.flyingbrick.de/abfrage.php
Select L Jetronic Motronic in the Suche in Teilgruppe box. The left column shows the BMWpart number, the column to the right of that shows the replaceement part number, and the far right is the description. You might be able to get Google Translate to help you with that or if that fails I can help.
-
Well, I will most certainly take you up on your offer as things get going.
Out of curiousity, can you further explain your concern with "water resistance" on the K mod you had developing? I have as mentioned, a long running MicroSquirt system commissioned on my 83 R100RT, and the ignition components (crank & cam chopper wheels and VR sensors etc. etc.) are all located at the very front of the engine timing chest and in direct line of fire to road spray and frontal rain blasting. I have been through some significant deluges and horrendous West Coast rain storms - never ever had a problem. In fact been on long distance trips about the continent with nary an issue (touch wood).
Perhaps something else you unearthed for the K, but my experience is better reliability than stock (had a trip spoil after my brother lost the Hall in his airhead bean can in Oregon and we could not seem to locate parts & trailered her home)
I was using a microsquirt module contained in an enclosure, rather than a sealed microsquirt unit. Hindsight 2020, I would have used the sealed unit, though the module provided me with some flexibility and customization, like direct access to the TTL level serial output for built-in bluetooth.
-
Question for you, as although I originally felt I'd wait till next winter to tinker on this, I have some time now to test things out in the workshop ahead of time (I like to run tests on tone wheel/ sensor combinations to prove valid cranking signal and low rpm snyc.
In any case, did you end up running your MS system with ignition control, if so did you try the Daihatsu 3 cylinder mode with cam tone wheel? I suspect not if you were tinkering with a crank mounted tone wheel as shown in your post. In any case, interested if you can share.
Lorne
-
I only ever did fuel only. The stock system is kinda unique, I had discussed methods of dealing with it on the forums. In general, it is a method that is hard to implement with the microsquirt system.
I had been looking at making a toothed wheel and hall sender setup, I've got a solidworks file I had made for it, attached below.
-
That is generous of you, I can have a wheel cut through my Son who has access to CNC equipment ( my lathe and milling machine are analog)
However, had you researched the crank mount location (my favourite by the way) to know if that option is code supported? I ask as the current information suggests only one 3 cylinder engine format supported with code and that utilizes a 4/1 cam driven wheel and sensor arrangement.
Although this is not generally felt to be as uniform as crank derived signals, I suspect the K75 cam gear train is pretty solid and doesn't create a lot of slop for a uniform rotational reading.
Regardless, if that is the only code supported setting then that is what I will start testing.
-
If you have all that available, it should be a cool project.
I had done research on the crank wheel support, and the problem is that the crank wheel on a k75 is actually essentially a 3-1 wheel, with 2 hall sensors allowing it to impute the missing tooth without having to wait. I think the stock bosch system uses the point where both sensors have a rising or falling edge (forgot which) at the same time to sync with (i think) cylinder 1. I considered using some logic circuits to convert it to an ersatz 3-0 and 1 tooth dual wheel setup. But I'm not sure you could code it to be accurate with the megasquirt methods, since the megasquirt really depends on counting the teeth to sync the spark.
One other option I considered is to make a timing cup that fits in the existing location with more teeth. One of the guys who did a K75 turbo did that, and used an optical sensor for what I think was a 12-1 wheel. I also considered modifying a stock K100 wheel as a starting point with the stock hall sensor, but those sensors aren't capable of using super small teeth, and the most I could have done was probably a 6-1.
-
Appreciate you taking the time to discuss these "esoteric" developments.
In the short time since I've started to research the K75 and how it might co-exist with an MS based control unit, I have determined that there does appear to be only one 3 cylinder engine type supported by existing code and tone wheel choice. That is the Daihatsu 3 cylinder with a 3+1 Cam trigger.
This I understand is a relatively thinly if at all tested format so I may be paving a pathway in more ways than one.
Because it is so difficult and deflating to test components on a running bike (or one that won't run properly or quits on the road), I will utilize the "JimStim" stimulator rigged on a test bench with ECU and lap top to prove sync, then I figure the best way to replicate a 3+1 tone wheel & VR sensor behaviour is to mount them to a dummy chain drive sprocket (introduce some jiggle) and see how she behaves at various rpm's.
More MS manual review in order, I'll be posting about things over on Project K Bikes
-
I'd definitely stick with crank based signaling, rather than cam. If you want to do full sequential, you can use a hall sensor in the valve cover, and pick up one of the cam lobes for synchronization.
-
You won't hear anything but agreement from me on the wisdom of the crank location for a tone wheel plus utilizing higher numbers of teeth to improve resolution and accuracy.
However (& I'm currently raising these points for discussion and debate over on MSExtra forums) there is a significant lack of 3 cylinder tone wheel and code options for ignition control ( & more importantly, sequential ignition) so we'll see what facts present.
-
One thing to keep in mind is that you can treat it as a 6 cylinder wasted spark setup as well. It's a little hacky, but opens up some options that otherwise aren't there.
-
I am no stranger to "hacky" contrivances - but have developed a safety net overview stance to redo anything that can be significantly improved, simplified, or take advantage of previously proven design.
Saying that, I've scored an earlier firmware version from a bloke in Norway that recognizes a 36 tooth crank wheel while doing wasted spark for three coils. I think once I'm back home I'll set up a lab and test that and a couple of other formats. I wish to nail the ignition component design befor I start focusing on other project aspects ( like turbo mount and manifold and plumbing items.)
-
:oops: This will prove to be easier than I had anticipated (misinterpreted tech advise over on the MS forums into believing there were limited tone wheel modes for 3 cylinder engines, nothing could be further from the truth and a bit of discussion over there has revealed top level options).
I will impliment sequential ignition and fuel injection control, with a single toothed cam sensor, and a 36/1 tone wheel and VR sensor on the crank shaft.
I'll be posting updates and detail on a post I have going on over at "Project K-Bikes" entitled "95 K75 on Mod Table".
-
:oops: This will prove to be easier than I had anticipated (misinterpreted tech advise over on the MS forums into believing there were limited tone wheel modes for 3 cylinder engines, nothing could be further from the truth and a bit of discussion over there has revealed top level options).
I will impliment sequential ignition and fuel injection control, with a single toothed cam sensor, and a 36/1 tone wheel and VR sensor on the crank shaft.
I'll be posting updates and detail on a post I have going on over at "Project K-Bikes" entitled "95 K75 on Mod Table".
Awesome! Are you going to be doing stock MS or MSextra?
-
I like MS Extra as it has lots of additional control for things like blow off valve & wastegate as well water meth injection. That means utilizing every spare circuit on the little MicroSquirt but has been working well on the R100RT. The last chapter I suppose would be a venture into Canbus control (also available)
So, basically enough to keep me entertained for some time 😎