Turbo Cooper - floor repairs
Turbo Mini Forums

Cool tread. You give me courage again for my turbo engine.

you can do anything if you set your mind to it...
i rather blow it up winning than keep it together losing.

finish date set for project 1... march 2018

I've spent way too much time distracted with this and trying to get it to work and although it's nowhere near complete, it pretty much does what it's supposed to, so I'm calling it done...

Built a dynamic balancer. I went through half a dozen iterations and this is what worked. This afternoon I balanced my clutch well within 1 gram (was 15g out).



I didn't think that changing the clutch was that big of a deal, but several interesting posts here http://www.turbominis.co.uk/forums/index.p...tid=578526&fr=0
suggesting that much of the imbalances are down to the clutch parts not the flywheel. When I used trial weights (neodymium magnets), I also found that it was much much better to correct the clutch plate not the flywheel...



Also had a play at lightening conrods. I made a jig that looks much like the piston dishing jig I made a few posts back, and this was the results so far...



Not that great, but it was just a test, so I have a good idea what I need to do to improve it and have plenty more 998 rods. Probably best if I put this aside and just assemble my engine now as I keep getting distracted with things I could attempt to do !! Maybe next year I'll have another look, and maybe try wedging a crank too?

Edited by PhilR on 2nd Feb, 2016.

On 5th Jun, 2015 PhilR said:
Built a dynamic balancer. I went through half a dozen iterations and this is what worked. This afternoon I balanced my clutch well within 1 gram (was 15g out).



)

you can do anything if you set your mind to it...
i rather blow it up winning than keep it together losing.

finish date set for project 1... march 2018

That looks clever!

I guess the crank is already balanced?

The crank sits in v-grooves made from blocks of PTFE. These supports rest snugly against piezo transducers. When an imbalanced crank rotates it tries to rock the supports from side to side and transducers detect the pressure caused by this and create an electrical signal. At the moment, I'm just using this signal to trigger a 12W LED strobe to show me where the imbalance is. I can then add trial weights (magnets), or cut metal away and retest. The micro-controller and screen aren't really doing much in this version.

The crank was factory balanced but I could detect a small imbalance. Now that I've balanced the flywheel and clutch on the crank, the crank's own imbalance has been cancelled out. I reasoned that even a perfectly balanced crank can have a slight bend and a 1thou bend will cause approx 1gram imbalance at the flywheel's edge as it's rotated off centre... Therefore 2 perfectly balanced cranks would likely require a slightly different flywheel balance anyway, so little point spending too much time fettling the crank. I understand that some manufacturers don't even balance their cranks - they just balance the flywheel and harmonic damper pulley once assembled to the crank!

Last month had ups and downs. I've done so much work to get the 998 engine to the spec I want, but every time I complete one job, I find two more !

Nearly complete and I found a big problem. When I put the 12G940 head on the small bore the fire-rings crossed over some corroded areas on the block where waterway gasket holes had been. I took the Dremel out to see how bad the corrosion was, and in one area it went about 1.5mm deep If I get the block skimmed, compression ratio is way too high. Yet another set back.

This was pretty much the last straw. This build has been so slow and time consuming trying to do it on zero budget, I was thinking about shelving the whole thing or selling up.

Then, literally the next day, I got a cheque from the tax man as I'd overpaid my tax a couple years ago - Totally forgot about it! This is a break I really needed

Went on eBay, bought a 1275 that looked tidy (998 engine will be put aside for now).



Appears to be low mileage economy metro, but with a large valve head (?!) Might even have an engine running soon

Edited by PhilR on 10th Sep, 2015.

Swapping from 998 to 1275, I was concerned that the clutch would be on the limit. I measured the new flywheel against some old ones and it looked like the pillars were 0.35mm higher than they should be, so I planned to trim them down to make sure the clutch plate would be clamped down correctly. I trimmed the pillars with this:





When I started cutting, I quickly found that the pillars are actually sloped down to the centre by a tiny amount. I still don't know why so I chickened out and just took a small cut. The spring is a touch flatter now but I doubt I made much difference better or worse. I think I'll run it like this and build up an RTS as a back up.

While it was set up, I had the idea I could cut my centre strap.



Initial cuts looks promising so gave it a try. Nothing to lose by trying. Final cuts were within a thou of flat, but surface finish was not super.



15 minutes of lapping, and job done

Edited by PhilR on 5th Feb, 2016.

Through-drilled the mains (hand drilled of course ):

Edited by PhilR on 5th Feb, 2016.

Slowly getting there... ticked lots of little jobs off and engine is pretty much ready to be installed. Fingers crossed and it'll be fired up this week

Final spec is:

1275cc block with 11 bolt, oil cooling jets, centre strap
New Rover 21963 high comp pistons
29, 33 valves with port and chamber work
9.0:1 compression ratio
Piper BP270i camshaft
Megasquirt 2
TF035HM-12T turbo (42.4-36.1mm turbine, 37.7-49.0mm compressor)
Standard gearbox with 3.1:1 diff (4 pin)
RTS clutch with extended clutch arm





Dual lambda ports welded in with take-offs for pressure sensors. I'll be using 2 widebands pre-turbo and a third one half way down the exhaust. The pre-turbo lambdas will most likely be used with an extender / restrictor to keep them a bit cooler.

The aim is to confirm that each sensor sees the same manifold pressure and to see whether they can easily be used to compare inner and outer AFRs (ultimately without the pressure sensors).

Edited by PhilR on 13th Oct, 2015.

Looking good.

Is that a knock sensor on the front of the block ?

And are you using controllers that can read the LSU temperatures ?

You've certainly got space to extend them if they do get too hot.

I think you will be the first to try them direct on the manifold rather than sample chambers so it will be very interesting to see the pressures. Graham is set up to measure pressures at the sample chambers I believe and I've got tappings in the manifold, which I will later use to measure the pressures like you, but they are currently occupied by thermocouples for testing Jean's new stuff.

That sensor on the centre branch looks a bit unusual, as in too few wires for a wideband (unless it's the photo angle hiding some of them).

Schrödinger's cat - so which one am I ???

Thanks Rod.

Yes, it's a generic, ebay knock sensor - I've not done anything more than mount it as easily as I could. I'm not going to interface it with the MS2 just yet, but will breadboard an amp and do the initial tuning with headphones. Depending on what I learn, I'll take it from there.

The 2 pre-turbo lambda controllers will do sensor temperatures (14Point7 SLC Free + LSU4.9), although I've soldered them up to be as low profile as possible to fit inside the ECU case without the displays... Initially, I will read the AFR and heater duty cycle via Arduino analogue inputs, then push the values into the the MS2 tables via CAN bus. As long as the heater output is somewhere between fully on or off I can infer that the sensor temperature is being held steady (and perhaps even how the inner and outer temperatures compare?). I have the programmer on my desk, so I'll try a small change to the SLC firmware get it to output the temperature via one of it's free analogue outputs (or maybe just reuse the pin for the emulated narrowband output).

I think you previously asked Alan whether he could do I2C output on the SLC controllers?? I'm not convinced the answer is as straight forward as 'No, there isn't enough memory' as I've had a quick play with the firmware and think a good compromise may be possible. Let me know what you think.

The sensor on the centre branch - I knew you'd spot that! It's a 4 wire narrowband I used while mocking up the manifold mods.

Edited by PhilR on 14th Oct, 2015.

Phil,

Some interesting points here.

Schrödinger's cat - so which one am I ???

You should sell the build plans or a kit from this. I would be interested for sure...

you can do anything if you set your mind to it...
i rather blow it up winning than keep it together losing.

finish date set for project 1... march 2018

Engine is in and I got it running today.

HEARD THE TURBO SPOOL UP FOR THE 1ST TIME - F***ING AWESOME, COULDN'T STOP SMILING

Still lots of work to do in the engine bay, but now I can get everything else ready its MOT





Rammie, I Only did it to see if I could and even then, it ended up being the bare minimum to do the job and nowhere near as complete as I wanted - If I was making them, this would be a very early prototype or proof of concept! I gave up spending any more time on it as it was getting me distracted from actually building the engine and I was struggling to debug it without an oscilloscope. Maybe I'll comeback to it another year.

Rod, I'll let you know what I find out with the SCL firmware, but will probably be in the spring at this rate. Regarding the LCD screens, I don't think I'll attach them, but just use the outputs for logging.

Edited by PhilR on 23rd Oct, 2015.

Engine's now had it's initial set up and idles and revs up beatifully, but will need lots of mapping on the road. I've swapped the 680cc SPI injector (0280150682) for a 1055cc injector (0280150655) which is the biggest I could find. After some time spent measuring and setting injector characteristics it idles great, even at at 4 pulses per cycle - that's about 1.0ms total injector time (including 0.4ms dead-time).





This set up should be good for 130bhp at normal SPI's 15psi fuel pressure. At some point, I'd be interested in getting a new pump and pushing the fuel pressure closer to the injector's rated 45psi. It should be able to flow enough fuel for 200+bhp but I'm really interested in reducing the injector on-time at full power. I want to see whether it's possible to counteract the siamese fuelling problem with careful injection timing and just a single injector... Not something I'd be able to do with the current fuel pressure as currently, the injector will be open most of the engine cycle at high revs.

With the engine now running, I spent an hour today working out how much welding I need to do to get the car through it's MOT. Water has been getting in the cabin for a while and it's time for some floor repairs. It looked quite bad, but after digging out most of the rust I've got a good idea how much work it needs so will try to crack on with it over the next month.

Edited by PhilR on 5th Feb, 2016.

First job was to remove all the sound deadening and clean up the floor so I could get a better idea of what needs doing. I can't believe how much it weighs, there must be 4 or 5Kg each side?

Edited by PhilR on 11th Dec, 2016.

I added up the cost of all the replacement panels I need and nearly had a heart attack - I stopped counting when I reached £500. Although this isn't going to be a full rebuild, anything I do fix for the MOT is going to be done properly... no more patching up. I'm pretty sure I'll want to take the car off the road in a couple of years for a complete rebuild, so any rust that won't affect an MOT test will have to wait until then.

Partly due to cost, and partly because I want to learn more, I'm going to make whatever panels I can. These are the inner sills...











I've got another mini in a similar state, so it made sense to double up all the panels I make.

This is a raised pressing in the original inner sill. I think it designed to have some kind of carpet clip welded to it, that was deleted in later cars? There's four of them on each side and although they don't serve any purpose now, I've recreated them to keep it original.



This is the stamp I used...



... and the first test stamping. I'm happy with that. Prob not needed, but was easier than expected and it would just bug me even though they're hidden! If there is any chance that the car will one day be returned to it's original spec, it would be nice to have the details correct, but TBH not something I'm likely to do!



I had a practice at making the companion bins last thing today, so hopefully should have them done by tomorrow. and then still need to make a load of random patch panels.

Edited by PhilR on 5th Feb, 2016.

I got some second hand extended sills that just have a bit of surface rust and trimmed an inch off them them and sharpened up the bends to replicate the genuine ones.

I had a think about making the outer sills, but I don't think I could do it as they've got quite a curve along their length and for £15 for this pair, it's just not worth it.

Edited by PhilR on 2nd Feb, 2016.

Cant see any pics at all

Some are working for me some are not

The raised squares on the inner sill is a throw back to when they used press studs to retain the carpets/ mats. Going way way back, carpet in the Mini was a luxury lol, just simple rubber mats normally haha, how things have changed..

No recent pics for me either,, just a no entry symbol, lol.

Schoolboy error - Pictures fixed!

Cudos to you at making the panels yourself

Looking good Phil

When my friend restored my mini, the inner sills you buy were too short. So they had to make their own like you have to reach the floor.

Seems you've made a good start!