Ducati ST2 : Amateur Restoration Take 2

[Mickly]

Dunno what the heatings like in your garage.

The cold snap may well have made the inlet rubbers stiffer & harder to fit.

Maybe try getting some heat into them to make them more malleable & some silicone spray to ease assembly.

[Hairsy]

Thanks - I really appreciate suggestions. I'm learning all the way. They're actually reasonably soft - it's more about lack of space to get fingers & hands. However, I did find a YouTube video that mentioned the problem and there was useful info in the comments. The one that I'm planning to try next is to attach the rubber to the inlet before fitting the airbox. The suggestion is that, with the aid of silicone grease, I'll then be able to refit the airbox and then connect the rubber to the airbox. This has promise as there will be ever so slightly more space.

 

 

[Hairsy]

I managed to get the rubber boot in place. For anyone who might be looking in future, my successful approach was ...

1) Attach the rubber boot for the vertical cylinder (left side of bike to the inlet itself
2) Attach the horizontal rubber boot to the airbox
3) Apply a good smear of silicone grease to the airbox end of the vertical rubber boot
4) Put the airbox in place
5) From inside the airbox, slowly work work the vertical rubber boot into place
6) Spend a couple of days waiting for the skin on the back of my hands to heal
 

It's hard to see in this picture but it's where the yellow circle is.  You can see the other cylinder clearly but that one's easy - the airbox neatly drops on top of that one.

 

Off to the NEC tomorrow to see what people buy when they're rich and don't need to buy projects.

[Simon Davey]

Good to see your success! 

[Hairsy]

Good day at the NEC yesterday. So many lovely bikes there - and none of them seemed broken. I'd really love a new bike one day - although it will almost certainly never happen!

 

My new plugs have arrived and I've also ordered some very cheap OBDII cables that theoretically will allow me to connect the ST2 to my laptop and, running free software, allow me to monitor the ECU inputs and outputs. If I can make this work for under a tenner then it will be a huge benefit in helping me to narrow down the running problem.

 

However, I have a slight diversion as I spotted on Wednesday that the tank was seeping from the welds at the back of it. Only a couple of drops an hour but annoying and smelly. It's the bits highlighted in yellow:

 

 

There's no corrosion or visible damage. It appears to just be a 'feature' of the build quality.

 

I'd really rather not change (or weld) the tank so I'm attempting a repair using Granville Petro Putty Fuel Tank Repair. I've read a number of very positive reviews but also a few less favourable ones. My hope is that the bad experiences are down to poor prep and that I can avoid that.

 

The first step was to empty out every last drop of fuel and then leave the tank open to allow remnants to evaporate.

 

Then I needed to take the tank back to bare metal in the areas that I'm planning to apply the putty. This was a case of 5 minutes with a Dremel and then an hour with various small tools to scratch off paint in hard to reach areas and ensure that all the surface was scratched to allow the putty to grip.

 

 

Then a thorough clean - initially with Isopropyl Alcohol and then panel wipe. I suspect they do the same job but I really want this to work.

 

And then finally the putty. It seemed more interested in sticking to my hands than the tank but I saw a tip from the manufacturer to smooth it down with a wet finger and that seemed to get it to work better. But it was impossible to get the beautiful smooth finish that I'd have liked. And the instructions say to apply it within 2 minutes of mixing, which certainly adds to the pressure.

 

 

It's gone very hard after only an hour or so and should fully cure in 24 hours. I'll probably give it a bit longer, just in case. If it's successful then the putty can be sanded and painted - and the whole area sits below the level of the seat so isn't visible.

I'd never come across this stuff before so have no idea what to expect. It may prove to be an embarrassing failure or a fantastic discovery. Time will tell.

 

I'll test in due course and report back ...

Edited November 22, 2024 by Hairsy

[Mickly]

Looking forward to the results

[Hairsy]

The putty on the tank has dried rock hard. I'm optimistic but that's not the same as being confident. The tank will remain completely dry in my garage so I'm not going to bother priming the putty and sanded area until I know the fix has been successful.

 

I have a slight practical issue right now as my fuel can is full of the fuel that I took from the bike. This includes the final dregs from the tank which will inevitably have some little bits of debris in it. I don't want to simply put that back in the bike. I also don't want to throw away 5 litres of fuel that is fresh. I've got some paper fuel filters coming, hopefully tomorrow, and my plan is to empty the fuel can into my car but through the filter. I'll then put 5 litres of completely fresh fuel into the bike - bikes will always get the best fuel!

 

So while waiting for that I've fitted the new spark plugs and air filter so that when I get the bike running I know that any issues aren't due to those. I also finally fitted the new battery that arrived last week. Despite no fuel, I gave the bike a quick crank to see what difference the battery has made and it's significant - it turned over much better.

 

While awaiting the fuel filters that will enable the next step, I turned my attention to the front indicators.

 

Each front indicator is held in place in the fairing using two tabs that are part of the original indicator moulding - and all four of those tabs have broken off. However, all 4 are still attached to the fairing. Goodness knows what hamfisted approach led to all four being broken.

 

Parts aren't easy to source but I've found a set of indicators available £60. However, although the tabs are in place on these second hand units, the lenses are in poor condition so I'm going to have a go at a repair. This is very much a restoration on a budget and £60 is significant to me.

 

The tricky thing with trying to bond tabs like this back in place is that if you don't get them in precisely the right position and orientation, when you later come to screw the indicators back on they will break again.

 

So my planned approach is going to be to tape the indicators into their correct location and leave the broken tabs screwed to the fairing. This should allow everything to be held in the right position before any glue, but there will be very limited access to actually apply the glue. So what I'm hoping to do is to use a small amount of epoxy to secure the tabs to the indicators in the correct position. I'll then let that dry with the indicators remaining in situ. That should allow me to then unscrew the tabs from the fairing and remove each indicator with its tabs in place. I then plan to build up the repair further with more epoxy and maybe a little fibreglass. I may be able to add some more material to the tabs and then redrill the holes which would increase strength. It will all depend on how everything looks.

 

Advice Request : if anyone knows a particularly good type of epoxy that will allow me to build up the tabs then that would be good. If everything was bigger then I'm confident that fibreglass could make a strong enough repair but with such small parts, there isn't space to extend the repair beyond the area of damage - which is how fibreglass would allow me to build strength.

 

Hopefully the fuel filters will arrive in the next day so, in which case I hope the next update will include a report on any improvement to the running that's arisen from the new plugs and filters.

I would be a bit of a surprise if that was my only problem - but a lovely one.

 

 

[Simon Davey]

Are those tabs acrylic? Either way, have a look at Tensol 12, it's pretty darned whiffy though, it's very powerful.

There's also Tensol 70, which I think is two part.

Edited November 25, 2024 by Simon Davey

[Hairsy]

Thanks Simon. I'm not quite sure what they are. They're just part of the original. ABS perhaps? I've just gone and taken another look and I think that, provided I can complete the initial gluing of the tabs on, there's a decent amount of space for more thickness. I've got some final metal mesh that I think could work really well for building strength - I'd basically use that in place of fibreglass matting.

 

I'll take a look at those Tensol products - thank you. It's the kind of job where I can try multiple products - if one doesn't work then I'll still have everything and can just try another. Although I must be careful not to spend more on trial and error than I would on replacing the indicators!

[Shepherd]

I had quite a bit of success with Plastex. After watching a couple of YouTube videos, I ended up piling the powder over the repair areas and dripping the activator liquid on to it. Sets well in not too long a time.

[Hairsy]

Another up and down day.

 

I started off with the first step of fixing the indicator tabs. I remembered that I have some JB Weld PlasticWeld so I'm going to try that first. If that fails then I'll look into the suggestions - a quick search on Plastex looked very interesting.

 

My idea of initially gluing the tabs in-situ was just too difficult - I still hold out that it was a good idea but trying to get everything aligned proved beyond my level of patience. Instead, I applied a small bit of plastic weld to the joining surfaces of the tabs and the indicators and then used a bit of tape to hold the tabs in place. I'll give that 24 hours and then look at building up the strength of the repair. PlasticWeld is supposed to cure in 1 hour but I'm in no rush so every application/layer is going to be allowed 24 hours.

 

The fuel filters arrived so I emptied my old fuel into the car. As expected, the filter was needed and did a good job. I'm glad this didn't go back into the bike.

 

 

Then I pulled another 5 litres out of my day-bike (which is now showing empty!) and put it in the newly repaired tank - and left it for an hour.

 

Today's good news - the Petro Putty tank repair appears to have worked - no hint of any leakage at all. I'll reserve my celebrations until it's been sat with fuel in it for a few weeks but I'm very pleased so far.

 

Next was to start the bike and see if the new plugs and air filter were all that was needed to solve my running problem. Unsurprisingly but disappointingly ... they weren't.

 

If anyone would like to have a go at suggesting solutions, here's a bit of video of it running.

 

 

It runs reasonably well when cold. As it warms, it needs to be held at around 2,000rpm+ but does sound OK. However, every now and then it appears to completely cut out. It will then start immediately and run for a few more seconds. This really sounds to me like something electrical cutting out or an injector just stopping - but it seems to come to a complete stop and doesn't appear to be trying to run on only one cylinder.

 

What strikes me as odd is the way it appears to cut completely, like all power or fuel has gone, but then start again. All suggestions welcome.

 

I've got some cheap OBD cables coming, hopefully in the next day or so. If I can check for fault codes and monitor the ECU inputs and outputs, that may help me to spot an issue. If not, my current thinking is that I need to get the injectors cleaned as they sat for 13 years without being used. I'd also really like to find out the expected resistances of temperature sensors so that I can check those - I'll need to do some research.

 

Finally, I did a bit of cleaning. I need to clean the old layers of oil and grease so that I can see any leaks. More thorough cleaning will come but jobs like this make me feel better when it's not going to plan.

 

 

[Simon Davey]

Quite a difference with the cleaning!

It actually sounds really nice, I wish I could help.

[Hairsy]

Yet another day of mixed success.

 

The initial gluing of indicator tabs had been successful so I sanded back all excess glue so that the surfaces were keyed. I then cut small pieces of metal gauze to cover the areas of the join on both sides. Having done that, I applied a layer of JB Plastic Weld Epoxy over the gauze. This is now drying but I'm becoming increasingly confident that this is going to work. Once it's dried, I'll need to re-drill the holes (which are now filled with epoxy) and sand back a little of the excess so that the tabs sit on their mounts. Hopefully tomorrow.

 

 

I checked the fuel tank again and still no hint of any kind of leak. So another bit of good news.

Then the postman arrived with the interface cables to allow me to hook the ECU up to my laptop and, perhaps, get some clue as to what's wrong with the engine. There are two cables : the first converts the proprietary Ducati OBD connected to standard OBDII (including a 12 v supply). The second connects OBDII to USB.

 

The connections, from left to right are :

 

Ducati to OBD Cable

 

1) Standard OBDII connector

2) Ducati OBD connector

3) 12v +/- supply

 

USB Converter

 

4) Standard OBDII connector

5) USB connector

 

 

And the problem is ... both of the cables have male OBDII connectors and therefore can't connect. Confession time - it didn't even occur to me to check this.

 

I can't open up the OBDII connectors. It may be possible to get a couple of female connectors and make a back to back converter but, for simplicity, I've now ordered a female OBDII to USB cable which should arrive at the weekend.

 

So the agony is prolonged - am I going to be able to get any useful information from the diagnostic port that will enable me to identify the running problem? I really hope so, otherwise I may have no choice but to start firing up the parts cannon. And I hate that.

 

I hope to finish the indicators tomorrow. That would be nice.

[Hairsy]

A few little updates today.

 

The indicators are now fitted and the fix appears to be good.. I think I've achieved a strong bond with the combination of mesh covered in the Plastic Weld epoxy. I used some additional washers as spacers when fitting the indicators to the fairing so that the tabs aren't under any undue stress. The real test will be when the bike has been used a bit and suffered some vibration but it feels solid at the moment.

 

 

While I await the OBD cable, I gave the bike a run up to temperature to check a couple of things.

 

Firstly I wanted to see whether the running problem gets better or worse at any stage during the warm up - this could point to a temperature sensor and/or perhaps a problem that is masked a little by the ECU giving a rich mixture when cold. I set the idle to around 2,500 (which is as low as it can hold at the moment) and just left the bike to warm from fully cold to fan-on. There was no difference at any temperature. While this hasn't pointed me at any specific cause, this could still be useful info later when I'm piecing together all the facts.

 

The other thing I checked, based on advice from a knowledgable friend, was what the battery voltage was doing while running. These bikes are known to have poor charging performance and it could have been that the problem related to the voltage dropping too low for ECU, spark or injector performance. However, throughout the warmup the voltage was a steady 14.4v. So that's a possible problem ticked off.

 

Finally, I can't remember whether I've previously mentioned it but there is oil leaking out from where the stator lead exits the engine case to go to the reg/rec. On closer inspection, it's been put together without the necessary O ring, so it's not terribly surprising that its leaking. I've ordered the O ring and taken apart the connector at the other end of the cable - this is necessary because I have to slide the O ring onto the cable and there's no way I was going to get it over the connector. The connector was a bit of a pig to get apart but I managed in in the end.

 

 

It's my wife's 50th birthday tomorrow so I suspect I won't be allowed to escape long enough to do anything post-worthy for a day or two.

[Yorky]

1 hour ago, Hairsy said:

It's my wife's 50th birthday tomorrow so I suspect I won't be allowed to escape long enough to do anything post-worthy for a day or two.

 Giving that sort of information out it could well be a month or two to allow for healing.

[Hairsy]

Couldn't help myself. My OBD adapter cable arrived today. I had no plans to run any diagnostics but I just wanted to check it worked. So I snuck out to the garage, connected the positive and negative for the cable to the battery and then connected the OBD connector on the bike. And immediately a load of smoke came out of the cable as it melted itself. Needless to say, I ripped it off quickly.

 

I thought I'd fried the ECU but I checked and the bike does start as normal.

 

On closer inspection, it appears that the cheap Chinese adapter cable wasn't wired up correctly.

 

I've now ordered a more expensive one and the first thing I'll do when it arrives is check that it's been wired up correctly.

 

So I think I got away with it but I think it was probably God's way of saying "don't play with bikes on your wife's birthday".

[Simon Davey]

Wow!

Thank goodness for quick thinking, well done for the save.

"Happy birthday darling, sorry about the charred remains of the garage" Probably wouldn't have gone down well.

[Nick the wanderer]

Re the not idling thing and cutting out, l think that may be fuel starvation. It starts well enough. Needs an Italian tune maybe.

If you get no faults when you get the obd reader set up that would suggest fuel even more to me, can't remember if you stripped and cleaned the carbs.

If you are gonna be firing any cannons it should be for the wife on her special weekend! Good luck, enjoy.

[Hairsy]

Thanks Nick. There's no carbs to clean so it's injectors that I need to clean. I've ordered a cheap cleaning kit. This, of course, is never going to be as good as a professional clean but I'd like to give it a try and see what happens. That assumes that I don't get an easy solution from OBD faults but I'm really not optimistic that it will be a simple code issue. However, if I can get the OBD to work then I may see high fuel trims where the ECU is trying to put more fuel in, which would validate the injector theory.

 

I'd really like to take the Italian tuneup option but the tyres are truly ancient and scare the life out of me. Also, I still need to go over the suspension & brakes. It's a bit chicken and egg since I don't want to invest the time and expense on the chassis when I haven't got a decent running engine. But it would help if I had done the chassis as I'd then be able to test the bike.

 

Fortunately, the aim of the project is to keep me busy all winter so there's not a great rush.

 

 

Today I fitted the O ring to the stator lead. It was uneventful but it still needs checking next time I run the bike. I also took the opportunity to put some additional heat shrink on the cables that carry the AC output from the stator. On close inspection, they weren't well protected. Shorting those cables could cause all sorts of bother.

[Hairsy]

My replacement OBD adapter cable arrived today. Before trying it I tested that every connection that should be there was correct and that there were no connections that shouldn't be there.

All fine - I connected it to the bike and this time there was no smoke or fire. A win!

 

Sadly it went downhill from there. I tried for over an hour to get my laptop to see the ECU but I just couldn't get a connection. There are a number of factors at play here:

 

1) The ECU & injection system are old '90s technology

 

2) I'm taking my necessary budget approach which means that I've bought cheap hardware that comes with very limited support. There's the adapter cable and then the OBD interface. The interface came with a CD that was piled with software and documentation but none of it is clear or easy to use and some isn't relevant. There are many settings in the software for ECU type, baud rate, etc and no hints at what I need. I therefore had to go through lots of combinations. I also tried multiple software options. However, ultimately, the OBD interface has 5 lights and only one is lighting up. There's no indication of what the lights mean but I feel sure that, with 5 available, only having one lit isn't a good thing.

 

3) On top of that, I'm running the ARM version of Windows on a virtual machine on a Mac M1 which is not a standard Windows setup. This has always been 100% compatible for me but there's always the possibility that it's adding problems.

 

I've contacted the vendor of my OBD interface one last time. His support to date has suggested he's selling on some units that he doesn't really understand though. I've asked for either help or a return.

 

I've found another supplier whose website suggested they really know what they're talking about and says that they're confident that they can get anyone set up and running. So I've emailed them with my specific setup details to make sure they retain that confidence - if they think they can help me then I'll order.

 

I could always just move forward and start cleaning the injectors but I still think there's a possibility that getting the OBD connection working could save me a lot of time and expense so I'm going to persevere for now.

 

Please be assured that, if and when I get success, I'll post up the hardware and software that I've used and hopefully save someone else the same pain in future.

[Egon247]

It can be a nightmare cant it? When i connected the laptop to my car, using the recommended software there was about 6 different settings that needed to be spot on before it'd talk the same language. 

Cleaned the injectors on my last car too, using the carb cleaner on a tube, then pulsed it using my benchtop psu. backwards and forwards. genuinely made a difference. Hopefully you'll get it in the end mate.

 

[Simon Davey]

@Hairsy

Could you borrow a Windows laptop? Just to rule out that virtual serial layer.

[Hairsy]

1 hour ago, Simon Davey said:

@Hairsy

Could you borrow a Windows laptop? Just to rule out that virtual serial layer.

Yes, I guess I could. At the moment though, the guy who sells the (probably) better version has offered to try to help me with support for the one I bought from someone else.

 

Successful or not, that makes him a good guy in my book - if anyone is looking for diagnostics cables then he'll be first on my list in future https://www.lonelec.com

[Hairsy]

Awaiting cables so a day of looking at fairing panels. Most of them are in pretty good condition but this one has a pretty big crack in it. I don't know what I'll end up doing with the bodywork but I wanted to at least make the panel structurally sound for now. I used plastic weld staples which I've used before and I've found to be very good and pretty easy to use. As a bare minimum, I'll apply some touch up to the outside of the crack. The crack will be low down on the fairing so probably won't be very visible - but if budget allows I'd love to get something more professional done with the tank and more serious damage like this.

 

 

The rest of the panels just got a good clean, polish and wax. There were lots of marks and scuffs but they came up really well. A nice satisfying job.

 

 

I learned a valuable lesson in the process. After initial washing of the panels, I wiped them down with paper towel and then finished drying them with my heat gun. After the last one, I inadvertently picked up up the heat gun by the hot end. I don't think I'll be doing that again. The label on the heat gun says it gets up to 500 degrees C. My hand's had 10 minutes under a cold tap and is now very sore. Be careful out there!

Edited December 2, 2024 by Hairsy

[RideWithStyles]

Sorry been away. 
just to note, sounds like the bike is trying to get off high speed idle (runs ok when cold) so injectors and pump seems ok, either not getting the signal it wants (electric or resistance) or when it does it’s not getting into the next phase of its program. When it does get off the high its struggles or stalls as its not being able to go onto its next bit of being able to run.

as a general rule a injector will either work or not at all! Yes they can get abit blocked or corroded but thats is few and far between. Those old injectors on the old ducs weren’t very good at fuel optimisation even at the time. The ecu wasn’t bad but the program and its parameters were not great…

yes you are correct some new programs and software (readers) might not work with old set ups, professional set ups are different, emulators are only as good as its design.