1987 Fiat X1/9 Turbo

Yeah that'd be dead cool only if you've got a spare 5 minutes!





Yeah they're the same base Fiat SOHC engine that was in lots of smaller Fiats from the 128 to the Mk1 Punto. As Duncan mentions they made a dizzying number of versions though. Closed deck/open deck, all sorts of different bore and stroke ratios that ended up making very similar (but not quite identical) displacements.

The early turbo ones were 105bhp at 0.7 bar, but they were intended to run at 1 bar (but insurance companies kicked up a fuss!). If you unplug the little boost sensor that cuts at 0.7 bar and add in a bleed valve to take it back up to the intended boost they'll happily put out 130hp and 145lb-ft.

This one's probably a little down on that, as we'll discover over the next few posts...





Yep! They do seem to rust pretty readily. Definitely one of the cars that built the Italian reputation for it. This one is in reasonable nick, but is still starting to go around the edges. Eventually it'll get a full teardown, de-rust and repaint but for now I've only got so many working cars so it'll have to soldier on for a bit longer!

And yeah, I can imagine they're pretty slow with the standard 1300 (and even the standard 1500 really). It's not just the outright power but the highish peak torque doesn't help them feel like they can scoot down the road. Something the turbo does help with immensely!



Thanks man yeah the second I saw it I knew I had to have it! When it came up for the affordable price anyway...

Onto the next update!

--------------------------

Running problems resolved! - 16th August 2019

So, during that whole time of rebuilding things I'd been chasing down intermittent running problems. One day it would run fine, and then the next it would run like absolute plop. This all started within the last 10 miles of the journey back from Wales. This was really rather frustrating as after all the work done to it over the course of the year it still felt like I hadn't really got anywhere. However, I do love a bit of a list and writing one out for the work done to date (rather than the work to do) did wonders for the motivation:

1. Removed dead nosecone and straightened panels underneath.
2. Fibreglassed on new nosecone and painted it.
3. Made new mounts for the front bumpers and number plate.
4. Knocked out the worst of the dents, did a spot of filling and painted them.
5. Bought alloy radiator from the States and made a tubular bracket to get it to fit.
6. Made stainless versions of the perforated coolant pipes.
7. Rewired the electrics for the starter and ecus.
8. Rewired the headlights so they have relays and don’t burn out the ignition stalk.
9. Replace burnt out ignition stalk from learning previous lesson.
10. Run an extra power cable from the battery to the fuse block.
11. Straighten front-left suspension links, and shrink the threads for the big top nut on the strut.
12. Replace both inner and outer tie rods on both sides.
13. Rebuild front and rear brakes.
14. Rebuild rear brakes again when they won’t bleed.
15. Rebuild rear brakes a third time with new hoses to get them to finally work.
16. Replace brake master when new brakes cause the old master to die.
17. Replace brake reservoir and hoses when pressure bleeder shoots a jet of brake fluid out of the reservoir .
18. Replace clutch slave when that blows a seal.
19. Replace clutch master when new clutch slave causes that to blow a seal.
20. Added a bronze bush to the gear linkage so it’ll select gears properly.
21. Modify an X1/9 sump and oil pickup to fit so it’s at the right angle for the car.
22. Replace coolant temp sensor and thermostat to see if it will make it run.
23. Try second hand AFM to see if it’ll run.
24. Replace fuel pump to see if it’ll run.
25. Change intake gaskets when I discovered I’d fitted one upside-down.
26. Modified a bung on the intake manifold to allow for easy removal.
27. Made a new set of HT leads, maybe that would make it run?
28. Tested fuel pressure and regulator. Both fine.
29. Checked for boost leaks. Nada.
30. Change rear coolant pipe that runs to the pump as it’s leaking.
31. Do timing belt and water pump.
32. Do valve clearances.
33. Modify coolant pipe as it doesn’t fit properly.
34. Modify coolant pipe again as it doesn’t seal properly.
35. Replace a second leaking coolant pipe.
36. Bleed the coolant system for the umpteenth time.
37. Finally replace the right sensor that was causing it to run like plop (rpm sensor on the flywheel).
38. Take it for MoT only for it to start running like plop again.
39. Spend another week chasing problems and rechecking sensor readings.
40. Miraculously starts running ok again (we reckon it was corroded connections to sensors that got cleaned when we were taking them on and off to test them).
42. New tyres (and slim spacers on the front because why not).
43. Troubleshoot some electrical problems with the headlights and dash illumination.
44. Refit a second hand idle control valve.
45. MoT and tracking.
46. Drive!



Scattered among that list is the stuff that I'd done to try and sort the running. Tested pretty much every sensor in the system, as well as replacing a fair few of them just to make sure (and because very little seemed to be improving things). The latest breakthrough at that point was discovering that the RPM sensor looked like this:



I replaced it with a new one which improved matters for a while but then seemed to have died again, prompting another replacement. By this point I'd finally started to notice a pattern. It tended to stop running nicely after bouts of heavy rain. Interesting...

Initially I suspected the RPM sensor again as I'd noticed there was some water ingress between the body of the sensor and the wire, but sealing that up on the new sensor didn't seem to help.

Then, about 17 months after it had first started running terribly...I cracked it!

I'd just fitted a second replacement water temp sensor as the first one I bought was reading funny resistances, but that made it run worse than ever. Then I remembered that it had chucked it down with rain the day before. Thought I'd check the plugs for water ingress. Made my way around the engine bay finding nothing, but when I got to the big plug on the fuelling ECU...full of water! Could have kept carp in it.

Little bit of searching around and I found this little drip coming out of the loom above the ECU.



What I think was happening was the drain holes in the engine cover were dumping water onto the mesh braid that covers bits of the wiring loom. That was then capillary action-ing all the way inside the loom, through the bulkhead grommet and spitting it out directly on the ECU plug!




I dried off the ECU and loom, wrapped up the braided sleeve in electrical tape so it wouldn't soak up water, rejigged the wires above the ECU so even if it did it wouldn't spit water onto the plug, and smothered the pins in dielectric grease.

After all of that, it ran! Halle-foxing-luya!

It still misfired off boost so there is something else going on, but great success! In a year and a half of driving after that it's never had those major intermittent issues again.

Yep! The PO when I bought it mentioned that it didn't like the rain, which I'd took to mean 'it'll rust away' at the time but in hindsight was probably a cryptic reference to the running issues.

Definitely glad I've sorted it as it's the sort of thing that could really ruin a long roadtrip (or even a short commute to work).

Awesome thanks! There's today's reading

I do think that 'David vs Goliath' type duel is one of the things you don't really get in the modern car world. Perhaps it's because there aren't actually all that many small sports cars about anymore, or that the power gulf between say an ND MX5 and a 340i is just so broad now (along with bigger sports cars not being quite as boat-like as they used to be, largely helped by truly colossal tyres).

Saying that, I did have a similar experience with a remapped VW T5 Transporter the other day. He was rather quick in a straight line, but from my perspective in the Fiat it was like he had crockery in the back aroudn the corners to be fair he might have!

About time for the next update I think

Gearbox swap part 1 - 21st June 2019

So, before I'd fixed the running properly I thought 'sod it, as it's not reliable I might as well do one of the other big jobs I have planned that would have it off the road anyway'.

There's three main things that need beefing up if you want to put more power through an X1/9: brakes, cooling and gearbox. The gearbox on this one was already showing the strain a little from just 130bhp (when it was running), and had started being pretty reticent to get into reverse...or fifth...or first...or any gear a little too quickly.

One option is to replace the internals with the ones from a Mk1 Uno Turbo gearbox, but as far as I can tell there aren't LSDs available for those. So, I've decided on the stronger C510 top-change gearbox from a Punto GT

The recent 10% sale sitewide on eBay netted me a quaife LSD so we're on!

First issue to overcome is the gear linkage. It's been sorted before using an MGF linkage so I figured that would be a good place to start. Here's the two side-by-side. The X1/9's is a solid tube so there's no good routing for that to the top of the gearbox, but with cables you can be a little more flexible.



The MGF linkage seems like a pretty good design. The only issue is it's bloody massive. The previous chap only shaved down the counterweight and bolted it directly to the tunnel, but that required a modified centre console so I thought I'd have a go at cutting it down to fit.



Started like so:



Ended like so:



This means that with a bit of minimal cutting to the centre tunnel it'll slot down neatly inside it





There's options to drop it even lower if necessary to clear the electric window switches, but that's basically it gearstick is in pretty much the same position as the original

Also, just look at all the extraneous weight I've chopped off

Car SOS X1-9 Episode tonight on Nat Geo

Gearbox swap part 2 - 10th July 2019

Did a little more chopping about to get the gearstick in the right place. If anyone's thinking of doing this, you don't need to go quite this far but it was nice to have some room to work with.



At this point in testing I discovered that the standard window controls which need about 70mm of clearance below them, and there's only 35mm of clearance between the centre console and the top of the tunnel, meaning you need to mount the linkage 35mm below the top of the tunnel to clear them.

You could mount it quite low down and a bit further forwards than the standard location which would be doable with the switches...but then the clamp for the cables fouls on the heater matrix return hose:



I was all ready to chop the metal line and run some rubber hose along the opposite side where it'll clear when my mate wanders in and says 'could you not move the switches?'.

Oh yeah. Sounds obvious now you mention it

Tested them in a couple of blanks that sit below the hazard warning switches and we reckon they'll fit once you remove the dividing piece of plastic. Perfect

So the gearstick can sit a little higher to clear the coolant pipe, and back in the standard position. I also decided to weld up a bit of a beefier 2mm mounting for the linkage as I'd noticed the 1mm wall tunnel had started to crack around the old mounting position. Probably not strictly necessary, but seemed worth the effort and added weight.



With that I had the front end hooked up (but if anyone's reading this far, make sure you make it to the end as I had to change a fair bit!).



Had to shave down some of the rotating part of the mechanism to clear the centre console in 5th and reverse.




Saved another 170g though if it flexes too much I'll have to put some stiffening ribs in, but we'll see how it goes.

Next problem though. Not sure how the previous guy got them to work, but the MGF cables are definitely too long for any orientation I could get them in.



They might have worked looping under the gearbox with quite a tall selector, but I didn't like how it all sat.

While I was puzzling through that conundrum I got on with fitting the gearbox (after having taken it apart to fit the quaife unit, which is the first time taking a gearbox apart!)



There's a bit of cutting to do of the bodywork and a bit of the gearbox to make clearance. The top section helps when it's fitted, and the left helps massively when fitting the gearbox.




A few people seem to have used the original crossmember and engine mount with a little plate on two of the holes, but offering it up they seem to put a lot of torsion into the mount so I've bought some bits to make some poly mounts and a different crossmember.

By this time I'd put some more thought and a bit of experimentation into the cabling issue. Speaking to people who have done it before, they'd put the cables down under the sump and up the back of the gearbox which does work reasonably well, but the cables aren't quite the perfect length so they get quite close to some other components. I think we can do better with custom cable lengths based on the long cable. This was the plan I came up with



The bottom picture is the new cable end. The yellow is an M10 threaded sleeve which threads snugly onto the rubber of the MGF cable (I tested it with an M10 nut which held on pretty strongly so a 30mm run should be solid. If not I've bought some rubber-to-metal glue). That slots into an M16 threaded sleeve (pink) and is welded in place.

A 30mm (or better 35mm) M16 bolt (light green) is threaded into that which clamps it against the gearbox bracket (brown). An 8mm hole is drilled through the M16 bolt, and into that is pressed an 8mm hard nylon tube from an air line to provide a bearing surface (dark green).

The upper picture is the swaged on threaded rod that connects the cable to the gearbox arm (green). Onto that is threaded an M6 female rose joint (purple) and a locknut (red). The light blue is an M6 nut with a couple of tabs welded on that bolts to the first rose joint, and into which the pink male M6 rose joint threads with its locknut. The second rose joint then bolts to the gearbox.

The reason for having two rose joints is that the gearbox arm describes an arc as it moves, so the inner rose joint allows the outer one to follow that arc. Without this, the swaged rod would bind in the M16 bolt (or you'd have to widen the central hole through the bolt to allow for the movement which would make it wear oddly and wobble about). The stock cables get around this by having the nylon tube that supports the rod swivel at its base.

Overall, the double-linked rod is only a couple of millimetres longer than the plain rod (could be less if you had half-nuts for locking, but I'm not sure how good of an idea that is).



It could probably do with a little more support through the 40mm travel of the rod, but it's still supported by at least 10mm at the furthest part out (which I'm not even sure it'll reach in normal operation).




That was mark 1 of the plan at least...

Gearbox swap part 4 - 8th August 2019

In a rare turn of things working out really nicely without me having to muck about with it, the MGF dust boots I had fit neatly over both the rose joints and the 24mm bolts. Just added a small cut down washer so they could hook the other side, painted everything and put it all together looks almost factory, save for the rose joints...



I also chopped off the little ball from the underside of the gearstick and rewelded it so it points back slightly (as it does in the stock linkage) and the gearstick sits nice and central now too




A bit of faffing around with cutting and epoxy-resining the console switch holders to relocate the window switches and I could put the interior back together you'd never guess much has changed!



Alternator's sorted with a bit of cutting and welding, and the Punto starter is one of those gear-reduction ones and is nearly 2kg lighter which is a nice bonus between this, the slightly lighter engine crossmember and the lightweighting on the gear linkage the whole swap's only added about 1.8kg, even with an LSD, stronger gearbox, and heavier equal length driveshafts (which you fit by using 2x X1/9 short driveshafts and a hollowed out stock CV joint used as a spacer).




Bit of a problem getting one of the suspension bolts back in as well. Previously it went in from the front, but the gearbox now fouls on it and the exhaust fouls on the rear. I took one look at the bolts on the turbo downpipe and decided to chop the exhaust and weld a joint into it so I won't have to drop the exhaust each time I do the bushes.



Does also make it easier to experiment with slightly lairier 'silencers' which is nice

Next up was that the slave cylinders you can buy nowadays for PGT gearboxes are all 25.4mm bore. Oddly, this flat-out wouldn't work with this setup for some reason as the biting point was pretty much at floor level. After much faffing about trying to convince people that 25.4mm bore slave cylinders aren't suitable replacement for 20.6mm originals, I found that a Fiat Doblo uses the same gearbox but has a 20.6mm bore slave. Only issue is that they come with a weird quick-disconnect hose fitting, but some drilling, tapping and filing flat sorted that out




Did the maths and if the 25.4mm slave moves the arm 14mm and that's about where the biting point is, the 20.6mm slave will move the arm 21mm so the 14mm biting point is 2/3rd down the travel which is pretty much perfect

Oh, and it's alloy too so saves about 250g which is a bonus

To bleed the cooling system on mine I ended up finding a steep hill and parking it nose down so as the top of the engine was the highest point! I think that was one of the more successful interventions I ever made on that car!

Haha yeah I've grown quite adept at it by now! I think I'm on my 10th coolant drain and refill after all the fettling (some more of which is coming up).



Thanks man glad you're enjoying it!

More cooling issues - 25th November 2019

So the fans and reduced leaks improved matters. It was no longer overheating as you were driving along on the motorway, but started to climb coming off it. It would cool down pretty well once you were up and moving again, but would get rather hot while sat at lights. So, I started suspecting that the water pump isn't flowing enough at idle speeds to keep the engine cool. One possible explanation I thought of was that the water pump end-clearance might be off with gasket+sealant+manufacturing tolerances.

Pulled the pump off and found the clearance was probably a bit big. Difficult to measure as the surface is curved, but the Haynes states 0.8-1.3mm and it was between 1 and 1.5ish depending on which vane I was measuring.



So, did away with the gasket and went for just sealant, and ground the vanes until there was a 0.5mm clearance on all of them (there was a good 0.6mm difference between the biggest and smallest vanes beforehand).



Might be a little tight even with the sliver of sealant spacing it out, but it doesn't touch even when hot. Quick test run seems to show an improvement, but it the engine temp would still climb a bit high before the fans came in when coming off a motorway run. At this point I'm pretty sure that the flow on the stock pump just isn't sufficient to keep a turbo engine cool at idle after good blast.

One thought I did have was to set up another temp sensor to run the fans that's on the engine end of things. That way the fans respond directly to engine temp, rather than radiator temp which is a good 4-5ft away from the engine. Or maybe an over-ride switch. Or just go all the way and get an electric water pump to either replace or supplement the mechanical one...

We have an afterrun pump on our S4, makes a noticeable difference to temps but we do also have 3 cooling fans as stock, 1 mechanical and 2 electrical. Turbo engines produce a lot of heat that needs to be dissipated.