chris846 wrote:Post #1: "The air flow meter is new me,"


What do you mean by this?

I'm not saying the AFM is defective. I'm saying that the tests might not actually be proving the injector circuitry because of leaks to ground through the way the AFM is connected possibly.

Hi Rob, crossed purposes: the OP described the AFM as ‘new to me’. Not sure if that means it’s an untested unit, or whether the technology is ‘new to me’. ‘Pologies if I’m off track, but the OP’s video shows his bike behaving exactly as it would if the AFM were disconnected - and he’s already mentioned damage to the FI loom....

chris846 wrote:Post #1: "The air flow meter is new me,"


What do you mean by this?

What I meant was that yes, the technology is new to me. Most of my other bikes have been carbureted and haven't had any kind of air flow sensor.

I reconnected all of the injector harnesses, and I did the "screwdriver test" again, and I am now getting rapid ticking on all injectors when running the starter. 
If there isn't a more likely culprit at this point, I'm definitely keen to look at the AFM as a potential problem here. As I recollect I cut/spliced the wires coming from it into the new harness. It would probably be worthwhile to open up the meter and test continuity to make sure that I spliced the right wires together and didn't short anything.

question to the group. 
Is it possible that improper throttle (idle setting) will convince the ecu to shut down?
as this was a very early version of Fuel injection for BMW motorcycles.

+1 daveyson computer stuff way over mine too.

Hi Trailkaus - did you look at the video I posted at #84? I can be a bit obtuse, sorry, but the video was basically me disconnecting the AFM on one of my bikes and then starting the engine - that coughing & half running that it does sounds an awful lot like the video you posted of your bike.
If you've had to repair the wiring loom to your AFM, I'd, well......fingers crossed but at least it'll be a cheap repair. Good luck.

Just to continue documenting: I pulled the EFI plug and tested continuity there to evaluate the AFM, as described here: https://www.kforum-tech.com/electrical/EFI/bike-wont-start-EN.htm
All of the AFM values appear to be pretty much as expected, with no indication of shorts or crossed wires.

@Chris let me go and take a look at that video

Paddy1 wrote:question to the group. 
Is it possible that improper throttle (idle setting) will convince the ecu to shut down?
as this was a very early version of Fuel injection for BMW motorcycles.

No, it won't. An improper idle adjustment will turn the idle switch off, but the engine RPMs will be less than 2000 at idle so the EFI is still in control of the injectors. The idle switch indicates either idle or full throttle. The idle switch in idle position tells the ECU to cut off the injectors on engine deceleration (you suddenly close the throttle while under way) and the injectors remain offline until the RPMs go below 2000 at which point the ECU again takes over control of them.

+1. The bike will run mostly fine without a TPS at all, or with a wrongly adjusted TPS. It will:

1: pop on the overrun because (as Robmack detailed) not shut off the FI when you shut the throttle quickly above 2000rpm.

2: Less obviously, have a bit of a loss of power/engine dying problem at low/mid-range rpm when the throttle is opened. This occurs rarely, but seems to require a particular combination of revs/engine load/throttle opening to expose this weakness - it's this combination that, I guess, is the reason for the WOT switch on the TPS - to prevent it happening!

I binned the TPS on the Madass, but I had to reintroduce a WOT switch to overcome problem 2 - it was dangerous with sudden acceleration (or not!) in traffic.

Good read this thread is!
Nice work guys on the fi relay!
At this point (based on last vid clip), I think your still getting over fueled. Did you check the temp. input to fi ecu. If so (and good) do you have access to known good injectors?

I have read some BMW bulletins on the early K's where the ECU, Ignition control unit, would get damaged because the case leaked and would allow water to enter. They had a range of serial numbers for US models but nothing for any where else in the world, so I don't know if this would apply to your bike. For the K100 0030001 - 0031222. K100RS  0040001 - 0041954, K100RT 0050001 - 0052267.

They updated the ECU in November 1984, to fix this problem, so I don't think it would apply to your 85, unless it had a 84 yr production date. This was a factory defect and the ECU was to be replaced on all bikes falling into the above serial # range.

I post this for info purposes.

robmack.
thank you for your reply to my question.

Hey everyone! I just wanted to swing back by and provide an update on this situation. 
Between my last post and now, I had explored a few areas in an effort to troubleshoot this bike. On the recommendation of a somewhat local vintage bike mechanic, I installed an aftermarket adjustable fuel pressure regulator, and dialed the pressure up to 50psi. This made no difference however. 
The pressure at the fuel rail was still falling off immediately when the starter was released and the fuel pump stopped running, suggesting that (since I knew the pressure regulator worked, and the fuel injectors had shown not to leak) the fuel was flowing back through the fuel pump, perhaps due to a failed check valve/flap. I wasn't sure if the lack of sustained pressure could be an issue, so I installed a check valve between the fuel pump and the inflow end of the fuel rail. The pressure now stays elevated after the fuel pump shuts off, however this made no difference in the overall problem being experienced. 

At some point I decided that the Electronic Ignition or the EFI computer were potential culprits and I just needed to spend the money to get replacements. Unplugging each, alternately, suggested that the Electronic Ignition was perhaps functioning properly and so it made sense to try purchasing a replacement for the EFI computer first. It arrived yesterday, I plugged it in, and the bike promptly fired up and idled without issue (other than perhaps being a bit smoky). Throttle response also seemed adequate, although I didn't test it thoroughly. I need to go back through the bike and make sure I've put everything back together properly. 

My current assumption is that the fire, experienced by the previous owner, melted through wire insulation in such a way that at least two wires were able to cross and short some portion of the EFI computer. It didn't completely fry the EFI, but did enough damage to cripple it. 

So now I have a Flying Brick that is, to be sure, better for the attention that has been paid it (even in areas that proved to be unrelated to the underlying issue). I look forward to having it out of the garage and on the road before long! 

My apologies for having taken so long to finally conclude this process, and my thanks for all of the input and advice provided!

Thanks for posting the solution. I found this to be one of the most baffling and interesting threads, and posting the solution really is the icing on the cake.

Don't throw your old fuel injection computer out just yet, it might be fixable. 

The test light results on my functional brick was constant earth to the injectors with ignition on, and pulsing earth with the start button pressed. Your results were no earth with ignition on.  I'm going to guess you won't have pulsing earth with the start button pressed. It should be pulsing, but so rapidly it almost looks constant, but you should be able to see it pulsing. 

If so, I feel the fuel injection computer simply isn't outputting earth to the injectors through pin 12. If you look at pin 12, it could be something as simple as a cracked solder at pin 12.

chase the wiring , and rectify the heat affected -melted cables , could be a cause for more failures in the future 

good to hear its a runner 

yippeee!!!

daveyson wrote:Don't throw your old fuel injection computer out just yet, it might be fixable. 

The test light results on my functional brick was constant earth to the injectors with ignition on, and pulsing earth with the start button pressed. Your results were no earth with ignition on.  I'm going to guess you won't have pulsing earth with the start button pressed. It should be pulsing, but so rapidly it almost looks constant, but you should be able to see it pulsing. 

If so, I feel the fuel injection computer simply isn't outputting earth to the injectors through pin 12. If you look at pin 12, it could be something as simple as a cracked solder at pin 12.

I just brought my K100 -85 back to home after sitting two years in garage (dry and warm). I tried to revive the battery, but it only kept ~7volts on it, so I bought new one and charged it full. I check the tank and it was almost full, but I didn't remember did I add any preservatives in it back in the days. The EFI was still hanging loose on the wiring harness (since I changed the battery), but I just thought that it doesn't hurt to at least quickly try to start the bike. Well the bike started like it was yesterday when I left it. I let the bike run on idle and after it was a bit warm I rev it and smiled like a kid with candy. Then in the mood I just thought that I throw the EFI back in place and once I tried that the bike died and doesn't start anymore. I checked that it gives spark and at least 2nd cylinder plug is wet. When I try to start the bike it sometimes "sparks" here and there, but doesn't start. I detached the fuel rail pipe and wet my self with fuel-->fuel pump works. I cleaned the EFI coupling and also opened the whole thing and didn't see anything burned or odd but cleaned it with electrical cleaning oil anyway and put back together...no help. I just ordered a fuel pressure meter to check the pressure, but I'm afraid that I burned the EFI...I'm familiar with solder iron and I've cheap oscilloscope, but I don't have much knowledge regarding the electrical components or how to search the fault. Also I don't like the idea to pay 200-300€ from used EFI that can be as broken or end of life as my own, so I would like to refurnish the one I've. I'll report what was the conclusion or next finding once some of the spare parts arrives. My budget is a bit tight atm, but I might need to sell the Wulf Gerstenmaier Special parts to someone on the forum if there is any interest. Also I can sell the whole bike if someone really wants

The EFI have proven over the years to be exceptionally reliable. You said the unit was hanging it was working OK but when you put it back into the case it started behaving erratically. I would be inclined to inspect the plug and make sure the pins are all contacting and electrically clean.
However if you do need a replacement it was found that some of the Volvo cars of the same era had the same Bosch EFI system and the ECU was identical. Some time ago I had a search around the net and found quite a few of them at very reasonable prices.
I am absolute crap at searching ( if I search for a horse I find frogs) but if you search the forum you will find reference to this ECU from Volvo.

I don't like my last reply, only open the fuel injection computer as a last resort.

The cheapest way is don't buy any parts until you know the problem. It's sat unused for a long time over the years, I'd give it a service, change fuel, oil, filters and the rest. That might fix it. The injectors might be blocked.

I think the fuel injection computer only has one output, pulsing earth to the injectors. Use the screwdriver test to check that all injectors are ticking while cranking. If one or more injectors aren't pulsing, do the LED test light test (described in post number 91) to find if the problem is an earth fail or a power fail. 

If you have pulsing earth to the injectors while cranking, I'll guess there's nothing wrong with the fuel injection computer. If you don't have a pulsing earth, the injectors aren't getting an output signal from the computer, that's a problem.

Thanks for the replies and help. I've good amount of service parts coming from all over the Europe and generally I'll try to change everything that's made of rubber and might affect. I'll change the fuel in the tank and replace the fuel filter+ all the hoses and vacuum line etc. I've gathered some guides all over the net and I'll check this forum deeper once I've the parts.  I'll post here the results after I've some progress and what I've done.