1997 VTR EFI conversion and other tinkering (short video P4)

[Nosepicker]

As said, very impressive work
Have you been able to ride it on the road yet?

[4x2]

I have the bottom of a Vfr tank with sender hole if you want it.

Thanks, but the bolt pattern is actually the same for the VTR/VFR fuel sender.
I might switch to the VFR tank at some point if only for the extra capacity.

As said, very impressive work
Have you been able to ride it on the road yet?

No, not yet, need to relocate the coolant reservoir (or make a new one) and it's way too hot to really get anything done.

[twinfreak]

Hey guys,
for me its not allowed to send pms because i am a New forum member....

If someone are interested at the red vfr800 tank with the inline fuel pump or the throttles from the sp1 with side plates for the VTR F:
Please contact me with an E-Mail:
Werner.lochmeyer@web.de

Thanks!

Twinfreak

[4x2]

Managed to do a very short (3-4km) test drive yesterday (I'll update photos later today, can't upload anything @ work)
Some observations:
- Tuning is obviously needed, the MAP load (vacuum) has a very limited resolution as expected with larger throttle bodies, so I'll see if I can mitigate this with closed loop lambda adjustment and some buffering of the MAP signal itself. If not I may have to switch to Alpha-N (TPS vs RPM) for the low rpm/load regions of the map.
- Battery is gone, oh well
- It didn't idle, which was a bit unnerving with the nearly dead battery
- Ducati throttle grip is stupid hard to turn for some unknown reason and the fast idle button doesn't do anything unless the cable has less than zero play, so I might just discard it and build some contraption for fast idle.
- Datalogging (via phone & bluetooth) worked quite well (tuning a bike on the street is a severe issue if this doesn't work)
- Power is definitely there, half throttle until 6-6500rpm it went well.
- Cam sensor seems to be leaking oil somewhere, not sure what material the valve covers are, I'll weld it if I can, if not I'll have to find something else.
Pro's: sequential injection/spark is a bit more efficient and hopefully it'll eliminate the backfires in the intake that happen during valve overlap (it sparks on both TDC's during the cycle, I suspect this *might* be a cause of the original carb farts as well)
Con's: More parts needed and it sometimes had issues synchronizing the crank & cam sensors during startup (so I turned it off for the time being)

All in all, not too bad, just need to find some time to flesh out a few of the hiccups and really start driving to collect data.

[Dickiebig]

Excellent news. Keep it up, keep us all posted. Most of this, tbh and in no way am I criticising your write up, went over the old noggin. I am a bit prehistoric and was at the back of the line when technical brain power was being dished out. However the more I read the better and less intimidating it gets. Not that I am about to junk my carbs

[4x2]

Excellent news. Keep it up, keep us all posted. Most of this, tbh and in no way am I criticising your write up, went over the old noggin. I am a bit prehistoric and was at the back of the line when technical brain power was being dished out. However the more I read the better and less intimidating it gets. Not that I am about to junk my carbs

Hehe, I assume if you're interested in trying it yourself you'd have learned enough at this point to understand most of the lingo, so no offence taken ;)

Updated photos as promised;
So, somebody's very, very, very happy with all my tinkering and it definitely isn't my girlfriend


So I kinda hate wiring, because I always wombles something up and my carefully planned cable routing goes to hell.
So far the VTR actually seems quite OK, but the rev counter isn't working yet, so I dunno what it'll look like in a few days.


And this is because couldn't be bothered to find a better solution, just made a 30mm long spacer that goes between the bracket & fairing, put a tyrip around the front mounting point and wedged the reservoir in there.
Worst part is that it actually doesn't look half bad and fits pretty well...

[Dickiebig]

Your right it does seem to nestle in quite well

[sirch345]

Invest some money into an old superbike -> check
Quite a few hours of tedious work and swearing -> check
Carb farts turned into throttlebody farts -> check...

But it's running

That is impressive I agree, well done

Chris.

[Dickiebig]

Absolutely, love it.

[tony.mon]

And this is because couldn't be bothered to find a better solution, just made a 30mm long spacer that goes between the bracket & fairing, put a tyrip around the front mounting point and wedged the reservoir in there.
Worst part is that it actually doesn't look half bad and fits pretty well...

An HRC reservoir fits in the same place very neatly indeed....

[MK_WF]

I was riding on the road and logging all the way.
Then at home feeding it into MegaLogViewer and optimizing only those partial regions of the map, I've been riding in.

Then burn the new map to the ECU and off you go for the next iteration.
In my case the hardest was to find the optimal AFR for each region. Funnily the low rpm & TPS range benefited from lean AFRs like 15.5 combined with very low ignition advance.

At the beginning I tried to cover all the map area by doing accelerations with constant throttle from very low up to max rpm. (constant throttle because I use AlphaN)
That gives you a base map pretty quickly.
Than later I focused on partial throttle and rode for longer time in the range of interest to gather enough logging entries.

Altogether it took ages but I could do everything on my own and didn't need expensive dyno time.

The result was a very smooth engine in that range :

[4x2]

That's actually how I intend to do it as well, find a quiet industrial area, drive up and down a bunch of times trying to get as much data as possible and have megalogviewer sort it out. Full throttle is usually very easy, make a run on a stretch of road in 3rd gear or so, analyze, adjust, do it again to verify and you're done within 2-3 runs. Low throttle cruising takes more time as it's less predictable and more difficult to get the conditions exactly the same each time.
Once it's running reasonably ok I can take it to work and do the analysing there (I usually don't have much to do when I'm not in the field)
For the time being I'll leave the acceleration enrichment off as well, just have to be a bit smoother on the throttle.

In other news: decided to look at the leaking cam sensor, at first made a new adapter plate that I wanted to secure with 2 bolts rather than 1.
Well, where's the fun in that compared to TIG welding oil soaked cast alloy on a 35degr day ?
So yeah, there I went...



Ofcourse I milled away the PAIR blanks a little bit too much ( ), resulting in a lot more welding to be done, but I'm fairly certain it won't leak there anymore.

And the other job was the tachometer, that didn't work at all, already changed the pullup resistor from 5V to 12V on the Speeduino, no effect.
First started to test the output while hooking up the speeduino to the function generator; seemed to be working just fine.
Tapping the tacho wire on the positive battery post resulted in the odd jump on the tacho, so that seemed to work.
Long story short: turned out the stupid thing required a signal voltage that was at least 1.5V above the supply voltage, so that was never gonna work with a 14V charging voltage....

Decided to go for the simple approach, put a 7809 (9V voltage regulator for small currents) between the supply wire to the tacho and it works great now (on the bench at least)

[Dickiebig]

Excellent, I am getting into this. Loving it, if only I had a clue about it all

[MK_WF]

I run an Ignitech on my VTR and the tach runs on the standard tach output (12V rectangular signal, rising from 0 to 12)

That's the signals I measured on mine with a pocket oscilloscope :
https://www.vtr1000.de/index.php?thread ... k-vtr1000f
(login to see the pics)

[4x2]

That's actually the signal I expected it to take, but it didn't, not even directly hooked up to the function generator & power supply.
Might be the tacho is actually kinda fried, but if it works this way, mehhh
The VSS it did take (although I did overguesstimate the pulses/km by a factor 3, so my little test run got me up tot 290kmh according to the logging..)

I'm using the DSO Nano as my pocket scope, very, very valuable tool (although a 2 or 4CH would be nice)