Friday, May 1, 2015

Ducati 888 SP4 charging system issue and influence of bad fuel pump.

I had a customer drop in this week with his 888 SP4.  The charge light on the dash had been coming on and staying on above idle, and the rpm required to make it go out was getting higher and higher.  A voltage check at the battery showed it struggling to get to 13 V at 5,000 rpm, so clearly something was wrong.

I did the usual checks, and with a rectifier connected directly between alternator and battery to check the loaded potential alternator output (unregulated), the most it had was 13.1 V.  Usually, you give it a quick rev, watch the voltage go to 15 - 16 volts then shut it down before the battery throws a hissy fit.  Sadly, 13.1 volts was not going to induce any hissy.  Well, not in the battery anyway.

I tracked down a new alternator at Frasers in Perth, which was air bagged over and fitted.  Unfortunately, the old alternator looked great, something of a bad omen as it turned out.  I ran the rectifier test again, and this time we had a mind blowing 13.5 V.  This made me rather sad, and very perplexed.  I tried another battery just in case, using one of my spares that had been charged the previous week.  You need to make sure you have a fully charged battery fitted when diagnosing charging faults, but even with this fitted there was no change.

The 851 and 888 SP use the single phase "narrow" alternator, the same as that fitted to the 350, 400, 600 and 750 engines pre 1998 (and 900 until mid 1992 or so).  The normal 851 and 888 use the "wide" alternator, also fitted to the 906, 907, 1993 on 900 and the later 748 and 916.  I had a stator only in a box with the wide part #, but I couldn't find the complete wide alternator I have (I saw it in a box a month or so ago), so I wasn't actually sure what was in the SP until the new one arrived and the old came out.  The photo below, which I stole from an Ebay ad, shows a wide rotor single phase alternator.

Turns out the wide stator is the same width as the narrow stator, with the steel part of the stator being 30 mm wide.  This fits with the Electrexworld listing showing a 30 mm stator (which again was confusing me).  But, the rotor is wider.  The narrow rotor is 30 mm total width, and the magnet section is 24 mm wide.  The wide rotor is around 40 mm wide I believe.  I don't actually have one to hand to measure (it's in a plastic crate somewhere), but it's something like that.

Given that I had a maximum of 13.5 volts, I figured I'd best go looking for the rest, if indeed any more existed.  This is the same alt used in a 600 or 750, and I have seen them go 14.3 V regulated easily, and much higher when running through just a rectifier.  I was struggling to think of places for the extra to go.  I tried lights on/off, which gave a steady 0.2 V change.  Most things will drop and then recover the voltage when you turn the lights on, but a drop and no recovery meant there wasn't anything more to be had.

The main difference I could think of between an M600 and an 888 SP4 is that the SP4 is injected.  So with it running and the tank raised, I popped the fuel pump connector and watched the battery voltage as it ran down to a fuel-less stall.  A 0.4 V increase to 13.9 V showed a potential worth pursuing.

This bike has only done 9,000 km in its 23 year lifetime, but the fuel filter was dated 06 from memory, indicating it was 8 - 9 years old at most.  The little round yellow filter on the bottom of the pump was damaged, with a few holes in the bottom section.  Given there was quite a bit of loose debris in the bottom of the tank, it had possibly been sucking grit even though the pump had a steel mesh screen on its entry.  The fuel filter (a KL145 in this 1992 model) was so blocked that I couldn't blow through it at all.  Along with the filter I replaced the hoses with some R10 spec and assembled it enough to try it again to see how many volts I'd found.  The result was maybe 0. 05, summarised as pretty much none.

At this point I was running it with the regulator connected again, so we had around 13.2 volts at the battery up at 4 - 5,000 rpm.

The SP4 has a "small" pump (not the larger pump of the 1992 and prior bikes), which is replaceable with the California Cycleworks FP-916.  Originally a Walbro pump, it does have individual electrical spade terminals (earth narrower than +ve), unlike the later style replicated by the FP-916 which has two spades the same wide carried in a black plastic connector block.  I cut a black plastic connector from a dead fuel level sender (never throw anything out!) and rigged it to work for a test.  Just turning the key on produced a pump sound very different to the previous, and firing it showed 13.6 V at around 4,000 rpm.  Switching the lights on gave the expected (hoped) drop and recover, and it would hold 13.6 V no matter what.

So I swapped the terminals on the pump power wires to suit those required by the black plastic connector and fitted the pump into its hole for the last time.  The 1992 models use the same rubber pump holder as the 916 series, but it goes into a round holed steel plate holder in the tank, with ridges moulded into the rubber holding it in.  It's a complete bastard to remove with little access behind it, and even my nitrile gloves were getting holed and filling with petrol in the end, which I really hate.  And the tops of my forearms, having been rubbed against the underside of the tank cap recess repeatedly, were quite red and angry.  But, finally, it was back together and I was much relieved.

The charge light was now not showing at all, even at idle, and a stable 13.6 V, while less than I like to see, was better than what it had when it came in.  The voltage drops in the loom between regulator and battery were about 0.15 V total on both +ve and -ve, pretty impressive for a 23 year old wiring loom.  I often see up to 1 V in bikes 10 years old.  

While the alternator itself didn't fix the issue, it was certainly partly to blame.  But it still was an expensive half fix.  And the impact of the fuel pump issue on the battery charge voltage is something I haven't ever seen before.  Most of these charging systems, with a good regulator and wiring, will knock out 14.1 - 14.3 V all day every day.  People like to give the old SP alternators a bad rap total power wise, but I was amazed to find it actually offering so little maximum output voltage in an unregulated test.  It really did stagger me, to the point of thinking there's still something I am missing.  But, I did learn quite a bit over the course of a rather frustrating day, and that's always a good outcome.

1 comment:

Unknown said...

While struggling with a 1994 bastard of an SS, a Monster and some 748 bits I find myself regularly reading your experiences and hints. This one is very close to an issue I need to fix this winter and I'm really thankful for the internet and your work.