Odd starting issue

[speedy231278]

Not currently an issue admittedly, however I'm just curious in case it ever happens again and might be something actually wrong rather than just 'one of those things'. On Saturday I went to Beachy Head. Not to chuck myself or the bike off, but there happened to be an airshow there, and a rather large triangular Cold War bomber would be in attendance. In typical fashion, I managed to get lost for literally hours (instead of 1 1.2 hours to get there it took me 3 1/2!!) and being a really hot day with LOTS of traffic, the bike got really hot several times, as is customary with this family. In clear air it cooled down at an acceptable rate as usual. It never reached the red line, and never has, although I wonder if the gauge actually goes that far? It was probably 80-85% of the way over, which I have seen a few times before.

Anyway, eventually after embarrassing detours around the back end of Brighton, Withdean, Pevensey, Hurstmonceux and Christ only knows where else, I arrived at my destination. While parking the bike, it cut out. To be fair, I *might* have stalled it rather than it decide to give up, but the net result was the same. However, clearly upset by my incompetence, it refused to fire up. My immediate thought was that it was a repeat of the incident I had in Guildford when the reg/rec had gone sick and wasn't charging at low revs which caused the battery to be flattened and not supply enough juice to keep the ignition running. The excessive heat for hoursa has toasted it, I thought. Having got off the bike and had a check, the lights and indicators worked perfectly, so the battery was fine. But, the starter merely gave the most pathetic of whines, like the battery was very low. In the Guildford incident, it wouldn't even do that, the relay just clicked or buzzed. I got my camera gear out ready to go and fins a spot to watch from, and decided to give it another go. No start, but there was a little more guts behind it. Wondering if things would get better over time, I waited another five minutes, thumbed the start button, and the engine turned over as if nothing had ever been the matter in the first place.

Now, I've heard of people suggesting some Hondas have issues starting when very hot, but nothing really more specific. I've never had this before, but to be fair while I did stop the bike a few times when I was lost, it wasn't necessarily toasty hot on the gauge at the time. The only time I've ever had starting issues is if the engine stalls a few seconds after starting from cold, and after waiting a few mins it's fine. Again, that's something I've heard a lot of people talk about. I'm just curious as to the cause in the case? Does everything being really hot somehow affect the output of the battery? Am I getting far too far-fetched in my usually far-fetched theories by wondering if a really hot engine causes the incoming charge into the cylinder to expand so dramatically that upon compression the start motor cannot overcome the force required to compress it? Or could it have been something even more random?

[krisztian_andre]

I used to have the same problem until I sorted my charging system (new battery, new electrical connectors to the rectifier, additional earth wire from the rectifier to frame earth point). The resistance of metals increases with the temperature. If the voltage was near some threshold it will be below that when hot maybe?

[magg]

If the fan is cutting-in at the correct temperature then the engine will not be running hotter than Honda intended. With the fan running the water temperature should not exceed approx 100 C i.e. when stopped in traffic, and should return to approx 80 C once moving again. Check the fan operation if the temperature continues to rise above 100 C. What is your nominal running temperature?

Slow cranking is an indication of insufficient current to the starter motor. As you have an NC35 it will be headlights ON always. This continual electrical load should be balanced by the charging system, at least at high engine rpm, however at low speed (idle) it is likely to be a net discharge of the battery. Therefore lots of slow running could leave you with a flattish battery, great technical term LOL. Having said that, IMO the wiring of the charging systems fitted to the NC30/35 is not be practice and with some rearrangement can be made to operate more efficiently, so that the battery does not become discharged by continual slow running. You can test my theory by measuring the battery voltage at idle rpm and at 50000 rpm, then connect a lead directly from the rect/reg -ve output lead (green) to the chassis and repeat the measure at idle rpm and 5000 rpm. What readings do you get?

[speedy231278]

Well, that's along the lines of what I thought, but the indicators flashed fine, and high beam was still very strong, and normally the indicators seem to flash slower or stay on with not enough volts on the battery and the headlights are dim. The reg/rec is a couple of years old, and so is the battery. There was clearly enough ooomph there to drop the relay, which usually just clicks if the battery is low. Funny thing is it only did it the once when hot, yet I stopped a few times to try and work out where I was. Maybe after all the slow running in Brighton there wasn't enough decent running on the way to Eastbourne (a few times there was a good couple of miles of traffic to filter though on the dual carriageways) top to top the battery up fully before I got back to being lost and in 30 limits again, or maybe the low speed side of the charging circuit isn't working as good as it could. I'll get the meter out on it next time I go back to Sussex and see what it's making.

[magg]

Battery voltage may be sufficient to make the lights etc look ok but the state of charge can be insufficient to supply high cranking current.

Regarding the rect/reg, recommend that heat transfer compound be used beteeen the rect/reg and chassis mount and the contact surfaces be made flat and clean.

It is not a low speed charging circuit problem as such, it is a combination of the wiring arrangement that results in an inadequate charge to the battery and the constant load of your headlights. IMO, even when these engines are running above 5000 rpm the battery charging performance leaves a lot to be desired.

[speedy231278]

If the fan is cutting-in at the correct temperature then the engine will not be running hotter than Honda intended. With the fan running the water temperature should not exceed approx 100 C i.e. when stopped in traffic, and should return to approx 80 C once moving again. Check the fan operation if the temperature continues to rise above 100 C. What is your nominal running temperature?

Slow cranking is an indication of insufficient current to the starter motor. As you have an NC35 it will be headlights ON always. This continual electrical load should be balanced by the charging system, at least at high engine rpm, however at low speed (idle) it is likely to be a net discharge of the battery. Therefore lots of slow running could leave you with a flattish battery, great technical term LOL. Having said that, IMO the wiring of the charging systems fitted to the NC30/35 is not be practice and with some rearrangement can be made to operate more efficiently, so that the battery does not become discharged by continual slow running. You can test my theory by measuring the battery voltage at idle rpm and at 50000 rpm, then connect a lead directly from the rect/reg -ve output lead (green) to the chassis and repeat the measure at idle rpm and 5000 rpm. What readings do you get?

I don't know the temperatures as the gauge doesn't have them on, which is something I've always found a tad annoying. Usually in summer the gauge runs somewhere approaching halfway to the red. In traffic on a hot day it will get 3/4 to 7/8 of the way there, and then go back to normal after say five mins in clear air. There must be a little more to this than just a flattish battery as I would have thought that a battery would not recover any charge when the engine is not running. However, if as k_a suggests the battery was a little low, and there was extra resistance in the starter circuit due to the heat, then maybe the combination was enough to reduce the power that got to the starter to the point where it was not enough to overcome the compression in the engine. It was definitely not very far into the flat side, as I've noticed on the odd occasion where the bike hasn't wanted to start and the starter spins freely that the lights start to dim pretty quickly, which is why I disconnect them to save a bit of juice between tires. Yes, the relay switches them off, but unless you turn the ignition off between tries there's still a tiny bit being used. I've heard of people bypassing the factory wiring for the charging circuit and running a direct link, so if there's a big difference at the reg/rec I might look at doing this. It won't be hard to do at the reg/rec end as the connector rotted years ago when I first had a charging issue that was caused by the connector being full of road crap and corroding one of the feed wires until the spade end disintegrated. The connector was toasted by the extra resistance, and did not survive having the remaining connectors extracted. I replaced the affected ones with new spade ends and connected individually. The reg/rec was replaced about 18 months ago when the old one died, from memory it was a genuine one fitted a few years before I got the bike in 2003. The one currently fitted wasn't directly from China, but reflecting on the price of £30 (cheaper now!) maybe it is. Currently listed here: http://www.ebay.co.uk/itm/Honda-cb400-c ... 4606c4533b

If the volts don't stack up properly when I test it next time I'm back, I'll have a look for a new one. As I've already said, the wiring loom side of things is already on individual spade ends, so one from another bike won't be an issue as long as it's a 5 pin one. I've seen specific years of R6 mentioned, is there any particular reason? More reliable? I know the genuine Honda ones have a reputation for frying, although the last genuine one on my bike lasted about a decade.

[speedy231278]

Regarding the rect/reg, recommend that heat transfer compound be used beteeen the rect/reg and chassis mount and the contact surfaces be made flat and clean.

It doesn't have heatsink compound behind it, however a previous owner fitted a substantial aluminium plate to go between the subframe and the reg/rec to act as a bit more of a heatsink, and every now and then I take the reg/rec off make sure that everything is nice and clean. On another thread someone posted a link to that funky cushion-cum-heatsink pad material, so I might try some of that and see what happens.

[magg]

The NC30 temperature gauge is marked at 70, 90 C and H, during open running it sits midway between 70 & 90. When stuck in traffic it rises to half scale which I assume to be about 100 and then the fall after the fan cuts in and then remains around half scale as the fan cycles.

You say you disconnect the headlights sometimes when starting your bike? How do you do that? Your headlights should automatically disconnect when you press the START button.

As for your rect/reg, same type fitted to my bike. Solid metal base makes for an excellent heat path, just needs transfer compound or pad, but pad i would think expensive and hard to find in such a large size. Unless the extra plate fitted by the previous owner a good thermal bond to the rect/reg and chassis, it probably is doing mote harm than good. The chassis forms the largest heat sink possible, provided the heat can get from the rect/reg to it. Hence the heat transfer compound.

[speedy231278]

NC35 gauge is plain, nothing apart from a white arc with a red line at the worrying end. When I say disconnect the headlights, I mean physically disconnect. The relay works fine, however in between tries the last time I had starting issues (dodgy battery I think) I removed the connectors because the ignition barrel was sticky and had a really annoying habit of not turning unless you had the key at exactly the right height! So it was easier to save the battery between starts by not having the lights plugged in that turn the bike ignition off and argue with the barrel. Gawd knows what was going on there, but one day it just stopped playing up!

[magg]

If the your fan is cutting in and stabilizing the temp before it reaches the red mark that is a good sign. To be sure it happens at 100 C, you could test the fan temp switch by placing it in a container of water, bringing it to the boil and noting at what temperature the switch closes.