Apologies if I've missed it but on this first failed cold start is it cranking normally but not starting or is it not cranking? If its cranking but not starting but starts second time ok and then starts first time the rest of the day, that is exactly what my Thruxton does often and its very common amongst the other water cooled twins. Sometimes it starts first time for weeks then has a spell of needing two starts from cold, doesn't seem to be any pattern to it, I don't believe its a battery voltage issue because if the voltage was too low first time it would be even lower second start when it starts ok. Someone suggested using "regular" unleaded over "super" unleaded fixed theirs but mine does it on both fuels, tried an extra fuel pump prime before starting but doesn't make any difference. Completely different bike and ECU I know and my Daytona has always started on the button first time every time.
Oh this is where the plot thickens and gets annoying, its almost random to what it does. Sometimes the bike will crank and have a tiny burp but not quite start, other times it will mimic the other starting issue where it will attempt to start but just not turn the starter motor, followed by going through the whole dash sweep again, or it will crank and fire and confuse me even more. What is strange however is if I wait for the headlights to come on because mine are on a delay with the dual headlight setup, it will crank but just not get over the line into starting, though it can also repeat the issues above if it likes. I should add it did this before I fitted that mod so I'm sure its not that anyway. After a number of attempts it will go but it all depends, it could be that the battery needs to warm up a bit because its a lithium, have tried a much higher CCA lead acid and the same issues persist, though that battery doesn't last as long on cranking until its dropped below being useable. I've also tried turning the bike on for a couple minutes then off and back but it does the same as above. If I hooked up the boost pack the bike will instantly fire into life, it's the only issue that I've got now, and after talking with the only two people I could find online that had done the same build, though theirs are track bikes without the extra electronics, the exact same issue was happening, if they boosted the battery it would fire and they'd not have a single issue for the rest of the day. However people who have put the engine into a gen 3 frame don't have this issue which points me towards the ECU change from the gen 2 to gen 3 being the culprit, but I'm not quite ready to change all of that just to fix a cold start issue.
That's classic lithium battery behaviour and needs separating from the other symptoms. Annoying that you have seemingly two different starting problems, no crank and dash sweep is normally classic low voltage weak battery fault but if it starts subsequently then instead of the battery I would look at a bad connection, faulty relay/solenoid or bad earth. Possible that using the booster pack gives enough extra power to overcome the voltage drop and allowing a start where it won't with a standard size battery. Regarding headlights, can't remember on the gen 1/2 but on my gen 3 they aren't on until the engine's running, and on any Triumph where the headlights are on with the ignition the headlight relay is supplied through the starter relay normally closed so the lights drop out while cranking to reduce load. With a cold lithium battery people find it helps to have the lights on first to "warm up" the battery, I would go back to a good standard size AGM at least until it starts reliably to discount the "cold" lithium battery as a cause. It looks like the Street Triple 765 and the Daytona have the same size battery (YTX9) so I'm guessing the starting loads aren't dissimilar and transplanting the larger engine in itself shouldn't cause low voltage issues. The cranking but no start problem but then starting second time (or warm) is, as I said, similar to my Thruxton and I think its maybe a fuelling issue, just not getting enough fuel of the correct mixture to fire first time after the bikes been stood a while and then the ECU aborts the process after a few rotations when it fails to start. That's a guess though and not sure why it would do it sometimes and not others if the system hasn't changed, maybe environmental conditions, and possibly again the booster pack (or a brand new fully charged battery) just gives it the extra starter RPM or even spark voltage that enables it to start first time. All a bit of educated guesswork though and probably no help, I haven't got any personal experience of transplanting a 765 engine into a Daytona or even a Lithium battery on a bike, but a lifetime of electrical fault finding.
It's all helpful in one way or another. But I should add, I've tried lead acid batteries and it had more power than the oem one, also tried an AGM one too, same issue was across all three batteries, that's why I'm not its not those. I do plan to check the ground point if I get a chance to play with the bike this weekend and see if that was causing it, I'm not sure it is tbf, which leads back to having gone through the long checklist of things to try based on what we all know from when the bike was a 675, the issue finding now is based on a setup and system that has next to no information on. The ones I do find that have done the swap and it all works are the ones who pulled over all the electronics from the engine side of things, which leads me back to what I thought of before, I even contacted a few triumph dealerships because I'd heard about lowering the voltage threshold on the battery, no luck whatever as it cannot be done on that ECU, but it can be done on the later ones. We'll see, the bike can be started and runs fine once its warm, so I'll take that as comfort in having a working bike while I narrow down that final issue.
Does it start second time after a failed attempt without any changes or the addition of a booster pack? If so then I can't see how its a voltage threshold issue as the voltage level will always be highest on the first start and then get progressively lower with every failed attempt, disregarding Lithium battery oddities. If the 675 electrical system has been retained largely unchanged with just the addition of the 765 engine then only an additional starting load could be causing a voltage drop, but they have the same size batteries as standard so I doubt that's significant and again if it failed on voltage threshold it wouldn't start second or third attempt. You've had failed cold starts with these batteries but not with a booster pack or have you only connected the booster pack after its failed to start?
Update on the bike, sorry I've been a bit quiet but I've been doing quite a few tests aided by AI which has involved a lot of back and forth conversations, so bare with me while I explain my findings. As I've stated the bike has a cold start issue, the ECU resets itself during cranking, this is often given a misleading view of "Oh its the battery, it does that" this is wrong, what it is actually doing is what I've now come to understand and learn is voltage sag/brown out, its what happens when the ECU can't react to a drop during cranking fast enough, it resets, it turns off, it doesn't log the fault because it happened so quickly, we've all seen it, we've all changed the battery and the problem 'went' away. A year later it would come back and we'd repeat the process. For my testing process I did a static battery test, recorded the figures, I then turned the key on and checked, I then pushed the start button and recorded my findings, so far these read.... Static battery - Key off - 13.6 Static battery - Key on - 13.45 TuneECU reading voltage - 12.5 Cranking - 11-11.5 So all these figures show that the range and dip is in the acceptable area for the bike to fire, I know it will fire if I attach a boost pack, what that boost pack is doing is forcing through a stable amount of voltage, it doesn't care about voltage sag, it gives the bike a stable feed and stops that ECU from browning out. As you'll also see, there is a one volt drop from what the battery shows to what TuneECU and the ECU itself are seeing, this is caused by that feed wire going through various sensors on its way to the ECU. If we take this as a simple math idea under cranking the battery drops 2-2.5 volts, if we think about this in plain terms that could mean the ECU reads even lower, 9.5-10 volts, this will instantly cause it to reset and brown out because of how fast and how sharp that drop is. But the drop is so sudden and so fast that the ECU is spinning around before it knows what the hell just happened. Now here is the kicker, how do you solve this issue without changing the loom, the ECU, the dash and everything else. It's an issue which has only come about from an engine change, or at least has appeared more easily to spot because of that engine change, that engine and starter require more, the old ECU and loom were designed for the 675 engine, they were designed with a small margin of error, what I've done in changing the engine but keeping the older ECU and wiring is expose that margin. The newer bikes corrected this issue by updating the loom, fitting different things in place, Triumph did this because those of us who had starting issues were given the false hope that fitting a new battery would cure all, which it did, for a time. Triumph didn't expect any of us to keep our bikes this long nor tweak and even in my case slap a 765 engine in and keep the old dated wiring and ECU What I've been doing is feeding my test figures and information on the build into AI, unlike us and myself who would search for information based on the make and model of bike I've found its been basing the information on the make of the ECU itself and has thrown up the information that our ECU's from the gen 1-2 were prone to weird resets, weird issues we could explain away with a battery but the issue could return down the line and we'd throw another battery at it for good measure. Now onto the possible solution which I'm waiting to do and hopefully this will correct the issue, a direct relay feed to the ECU itself from the battery. What I've done so far in relation to this..... 15amp inline fuse from battery to relay Single wire from relay to an ignition fed wire, in my case I've connected it to a wire on the tail light that only activates when the key is on Ground wire to the same place on the engine as the negative from the battery Wire from relay to the ECU live feed wire I had a slight mishap and connected it to the wrong wire using a posi-tap, no harm done but am waiting on some Y connectors so that I can cut the wire feeding the ECU with power in the key on state and run it inline with the wire that already feeds the ECU, thus creating a much more stable feed for the ECU to see. I'm not taking away the loom fed feed, I'm just adding an extra wire without all the noise. What I'm expecting and hoping to see is that with this direct feed vs the feed that travels through various sensors and sees a dip of 1+ volt from the battery to ECU plug is that the ECU should see a much closer voltage to what the battery has, so for example if the battery is reading static key on at 13.45 and the ECU sees 13.3 then I know it will be doing its job perfectly and it will remain stable during cranking, unlike what its doing now when all those sensors want a bit more during cranking which causes the ECU to dip further and reset. The boost pack eliminates the issue because it keeps the voltage stable, I've done 2000+ miles on the bike and new engine setup, no issues when warm/hot and no issues when its running so that rules out all the various sensors, coils and pretty much everything else, it's left me with one final piece of the puzzle to solve and I finally feel I'm closer, at least I hope I am. And that's the update so far, I'll get back to everyone once those final pieces arrive in the next few days and I can finally see if the direct relay fed feed does the job, AI seems to suggest that race teams have been using a similar modification for years so I can't see why it wouldn't work in this case.
Well you've certainly put plenty of work in, trying to resolve this Stephen. Fingers crossed that all that diligence will be rewarded with trouble free starting. Roll on summer.!
I think you have a problem with your wiring loom with those voltage readings, just checked mine and with a battery voltage of 12.63v Tuneecu voltage reading was 12.4v. You say the 765 engine requires more current to start (and therefore voltage drop) but do you have any figures to back this up, it uses the same starter motor and same size battery for both bikes so can't imagine the effect on a healthy battery is going to be barely any different to starting a 675 engine. When you say the ecu feed wire "goes through" various sensors, not sure what you mean, it may be looped or joined to supply other things but does not "go through" them, and aren't all the sensors on a separate 5v supply anyway? The length and cross sectional area of the wire and any connections will create a resistance and therefore a voltage drop, but this depends on the current in the wire (Ohms law), your ECU won't draw any more current than mine therefore your 1v voltage drop suggests to me a much higher resistance somewhere in that supply line to the ECU. Your relay "fix" could work but maybe only because you are bypassing the actual fault in the wiring loom, which could be repaired instead.
