Jump to content
  • Welcome to 205GTIDrivers.com!

    Hello dear visitor! Feel free to browse but we invite you to register completely free of charge in order to enjoy the full functionality of the website.

Sign in to follow this  
brumster

[engine_work] Engine Upgrade Mi16->s16

Recommended Posts

brumster

Figure I should record a few lessons learnt here as it might prove useful for others; there has certainly been a few minor issues I've discovered along the way.

 

Sort of follows on from this thread - http://forum.205gtidrivers.com/index.php?showtopic=81038 - which was the original mi16 conversion

 

Background

 

Getting to ~190bhp hasn't been too much hassle with the alloy 1.9 and I'd say it is cracking value for money and a really good approach for someone who wants to keep the alloy block, and keep reliability and power delivery to sensible figures and hence probably reliability. But in the search for a bit more oomph, it became clear that going 190+ on the alloy lump was going to get expensive. Addressing that 100cc deficit was going to mean custom liners and after talking with Longman it became clear that an iron lump was much more cost effective. Sure, you're carrying around 20 kilos or so more weight in engine block, but the gains should outweigh the losses - and to be honest a little bit of weight over the front might not be a bad idea at the moment, as I'm having fun putting down the power from the 1.9...

 

So I sourced an iron-block S16 from on here, budgeted for a rebuild and initially was going to do it myself. Once stripped it was clear it needed a rebore, and the shells were not ideal but the thing was workable. One thing led to another and Andy Baker talked me into just giving it to Longman and allowing their expertise to wash over the engine - Sid had built plenty to a similar spec and was confident of a 250-260bhp specification and with good torque (~180lb.ft) and for the aggravation of trying to sort the machining out myself with less-than-enthusiastic local workshops, I just said 'stuff it' and sent it all down to Longman :)

 

Anyway, this is more about the swap from an alloy 1.9 to an iron 2.0. Here's what I've found so far....

 

Engine Mount

 

I'd already tackled the engine angle issue with the previous mi16 conversion, so the hard mounts and maniflow 4-2-1 manifold coupled with the lowered radiator and throttle bodies mean the engine has dropped in at what, to my mind, seems an identical position to before. The central tunnel bit didn't need knocking in any further, presumably because my manifold is made for the job and clears all but the shell seam (which has been knocked flat with the previous engine conversion), so there's a good 10mm of clearance there.

 

However the engine mount itself is not the same as the alloy block, and the two are not interchangeable either. So make sure you get the engine mount that comes with the engine.

 

Ideally you want a non-aircon engine mount coupled with a separate alternator mount down at the bottom of the block. The gti6 engine mount, or a mount of any car that had air conditioning, is a pretty big beast designed to hold power steering pump, alternator and air con pump from what I can gather. It weighs a tonne and is large and overly complex, and probably wouldn't clear my TB's anyway.

 

I also wanted to keep my 8v alternator. Because I couldn't find the lower alternator mount in time, I ended up chopping that part of the casting off a full-blown gti6 mount, and then making up a small bracket that will adapt the lugs on the 8b alternator to fit into the narrower bracket of the gti6. Apparently if you get the correct alternator mount off a non-A/C 405/306, you won't have this issue as the alternator from the 8v will go straight on, but I couldn't prove that theory.

 

Oil Temp Sender

 

...isn't on the front of my block, there's only one hole tapped for the pressure sender/switch. I'm guessing the sender in the sump is the oil temperature sender. So just bank on moving your oil temperature sender around, or putting an appropriate T-piece in, if you need to. No big deal.

 

Oil filter

 

Again it would seem my block was from a car with an oil/water heat exchanger, which meant that there was no insert in the block where the oil filter screws on. Instead, there is a sleeved insert you screw in, over which the takeoff for the cooler sits. Since I don't have this setup, I needed to buy an insert to screw into the block - part #110354 - but didn't spot it until the last minute. Just be aware this can be different; worth keeping an eye on

 

Exhaust manifold

 

As some of you will know, despite the head being the non-RS head, you might think it was the same as the outgoing 1.9 head. Not quite - the exhaust manifold is located with 10 studs on my iron block as opposed to the 9 on the previous 1.9. Thankfully my aftermarket manifold caters for both stud patterns, so all that was needed was a new gasket, but it's something to bear in mind if you're not using an aftermarket manifold. Given the hassle around the exhaust manifold clearing the bulkhead, the stud pattern, the engine angle and performance in general, I think it worthwhile people just splash out on the decent maniflow design anyway - no angling plates, no hassle and possibly better power also?

 

Water pump takeoff

The little alloy casting on the back of the block above the lower engine mount - the takeoff from the water pump for all the various hoses - is not interchangeable between alloy and iron block so you'll need the original item from the XU10. No big deal, but worth noting. Thankfully the lower mount itself is a direct swap.

 

Thermostat Housing

Gordon bennet, the thing that came with the engine was a right load of gubbins! Loads of sensors, take offs, a cam position sensor built into it - just a very messy bulky piece of metal on the edge of your head. I basically junked it for the old simple housing off the previous 1.9 16v head

 

Clutch

I took off the old clutch, which was a standard OEM plate and cover for the 1.9, and was surprised to see that it was in remarkable condition. However I didn't fancy my chances on the new engine, so have opted for a Helix 6-plate solid clutch and cover, to handle the uprated torque. Pretty much a direct fitment to the XU10; nothing really to report on that one. Curious to see how heavy it is when it's connected up!

 

The engine is in but I'm waiting on a few bits to complete - the filter stud and the exhaust gasket for starters - before I can piece it all together and start it up. Then it will be time to map it up, run it in and see how it compares. Main things of interest are (i) the affect the additional engine weight makes to the car and (ii) how the driveline stands up to the additional torque, plus how well it puts it all down. I am suspecting I will need to move to 3-way adjustable front suspension next year to soften up the front and get some traction, as the current platform Billies still feel quite 'stiff'. The extra weight, as mentioned, might go some way to alleviate this though. So we'll have to see.

Share this post


Link to post
Share on other sites
DrSarty

Good luck with this.

 

You could use just Loctite 5920 to seal the exhaust manifold on. Many have issues getting a proper seal using an OE gasket and aftermarket manifold. The Loctite stuff is fab.

 

So you can confirm then that an XU9J4 coolant/stat housing bolts straight onto an S16 head yes?

 

Do you have an uprated diff? (Sorry: didn't read the original thread).

 

Did Longmans deck the block for up to 11:1 CR? (That's a 1mm deck, and you can go more if you're bold and valve clearance lets you).

 

16kg block weight difference apparently. Someone weighed the two (1.9 including liners) recently for my tandem engine build.

 

You're right about those monstrous "20,000 leagues under the sea" style alternator and pump mounting brackets. F*ck me. When I hacked mine about just to support an alternator rather than PAS pump it was a mere shadow of its former self, roughly 6% was left. It looked like an off-cut!

 

Isn't the oil temp sender in the sump?

 

Oil pressure and switch you're right can come off a t-piece, otherwise it can be one or the other. For some reason my S16 block had both...or am I being dumb, again?

 

The metal coolant union on the back of the block is much better, and has one less take-off IIRC than the plastic (s*ite) XU9J4 one.

 

GOOD LUCK!!!!

 

P.S. Hope you've got cleaned injectors.

Edited by DrSarty

Share this post


Link to post
Share on other sites
brumster
So you can confirm then that an XU9J4 coolant/stat housing bolts straight onto an S16 head yes?

 

Certainly does with mine, yep. Four long cap-head bolts on the side and that's all. It doesn't seem to locate on the head as such, ie. there's no dowels or a ridge or anything like that.

 

Do you have an uprated diff? (Sorry: didn't read the original thread).

 

It's a tran-x plate diff. Not sure how it will hold up; might just be swapping the plates a bit more regularly than normal :D

 

Did Longmans deck the block for up to 11:1 CR? (That's a 1mm deck, and you can go more if you're bold and valve clearance lets you).

 

Yes, it's been decked to suit but they aim for a lower CR than that, more around the 10.2:1 from memory, certainly not 11 or higher. I remember having a discussion with Sid about it, and the general summary was unless you were running expensive WRC-spec fuel the risks didn't warrant the rewards.

 

You're right about those monstrous "20,000 leagues under the sea" style alternator and pump mounting brackets. F*ck me. When I hacked mine about just to support an alternator rather than PAS pump it was a mere shadow of its former self, roughly 6% was left. It looked like an off-cut!

 

LOL, indeed yes, pretty sizeable chunk of cast, that!

 

Isn't the oil temp sender in the sump?

 

Yes, you're right, sorry I probably wasn't clear on that. Just that the block isn't tapped into the oilway on all of the iron blocks - certainly not on mine - more than once, so the oil temperature is indeed in the sump. I need to get another adapter plug to put it in there, or else fit a T up in the block takeoff.

 

For some reason my S16 block had both...or am I being dumb, again?

 

No, you're not - it seems some have two and some don't. Looking through the parts CD I can see some models have the oilway tapped for the temp sender, and others don't. Go figure.

 

P.S. Hope you've got cleaned injectors.

 

Oh they were brand new on the last engine build, haven't done enough miles to be dirty yet! Nice 305cc Bosch red'uns from that place in Oz. Wonderful place :D

Share this post


Link to post
Share on other sites
taylorspug

Interesting about the compression ratio.

 

Maxi is running 250hp through OE 309 shafts, no diff and 888s and all seems well so far, so hopefully driveline wont be too bad for you. Maybe consider some shortened 306 shafts?

 

As far as the weight over the front goes, i found when i went alloy to iron i had to up the front spring poundage a bit, along with adding some negative camber as the outside edges of the tyres were getting worked over more during cornering. Although my boot mounted battery has probably offset the block weight, so its all relative.

Share this post


Link to post
Share on other sites
James_R

You also need to block off the hole in the side of the head on the s16 head when using the Mi stat housing (it's the oil return for the vac pump)

Share this post


Link to post
Share on other sites
DrSarty

Yes: the CR comment now has me like a sump....baffled.

 

How can a decked block lower compression from standard 10.4:1 to 10.2, unless you're using the mutha of all head gaskets?

 

Even 11:1 CR on a mild road cam only needs 98 RON says PeterT, so unless you've added a super charger we don't know about I don't get it.... :D:D:(

Share this post


Link to post
Share on other sites
brumster

I might be wrong with 10.2 then, maybe it's high 10's? I know they didn't push it over 11 though, for that I am definite. There's custom rods and pistons in there too, so I'm not sure how that had an affect on things.

Share this post


Link to post
Share on other sites
brumster

Yesterday was spent making up and fettling this alternator bracket to adapt my XU9 alternator (originally off my 8 valve) to fit the XU10 gti6 engine mount.

 

The bracket in all its glory...

img_0794.jpg

 

Fitted to the alternator

img_0795.jpg

 

Fitted onto the chopped lower half of the gti6 engine mount

img_0796.jpg

 

And presto

img_0797.jpg

 

Fitted with a 698mm belt and it's just the job. The bracket itself is just a few bits of steel tacked together and braced, with a nut welded on the one side into which the cross-bolt screws through the engine mount and into. There's a little spacer (like a large sleeve) to push it over to the right as you look at it, and therefore line the pulley up. Works wonders and saved me a few quid (not that I could find the right engine mount anyway, so no real other option!).

 

Tomorrows job is match up the inlet manifold to the head, there seems to be a couple of mm difference on the ports on the head compared to the previous head I was running, so I guess Longman have taken them out a little. No big deal. Should be in a position to start her up tomorrow night, hopefully.

Share this post


Link to post
Share on other sites
brumster

Well, spent most of last night strapping everything back on.

 

Figuring out where all the oil pipes went on the front of the block was a little interesting, but thanks to the parts CD I could figure it out. I guess I'll write it down just in case.

 

The outlet from the head goes straight down the front of the block with a short length of pipe, and then enters back into the block via the little inlet tapped in just behhind the starter, up against the flywheel end of the block. The other two on the front of the block - the central, lower one and the short stub coming up out of the block behind the starter, go up to the filler. The filler has another outlet on it which seems to have gone to some sort of breather/valve thing previously, and I'm guessing circulated back into the air intake - so this will become a breather out to atmosphere (ie. my trusty coke bottle!)

 

Plopped everything else back together, popped some fluids in, and time to start her up. Cranked her up to pressure first, then on with the fuel and she started pretty much immediately. Unfortunately it was 10pm and I was knackered, so the drive will have to wait until the weekend.

 

I notice the top-end is quite 'clacky', which I guess is normal for a new set of race cams and solid lifters, and will quiten down with a few miles/hours running at 2k.

 

Big challenge is going to be the bonnet - the TB's are now a little high and proud of the bonnet line, to my eyes anyway. I really don't want to have to cut a hole and put a daft lump in my bonnet; is there an alternative performance manifold that will result in the TBs tipped downwards a little? Or should I resign myself to the inevitable and get cutting and moulding :-(

Share this post


Link to post
Share on other sites
welshpug

plenty of manifolds out there that'll have them at a better angle, probably more length from what I've seen of many, give Sandy or Colin a shout.

Share this post


Link to post
Share on other sites
brumster

I think I might do that. It's quite close, I've tried the bonnet on and if I lift it up slightly at the rear it's just touching on one of the re-enforcing webs - the filter backplate, not the TB's themselves (sorry, poor wording on my part in the previous post). I'm going to try cutting this web down tonight or over the weekend to see if it resolves the problem and if it does I *might* just get away with it..... really don't want a chavesque bump in my bonnet if I can help it :P but the car is more functional than stylish so it doesn't really matter, I guess. Be a shame to cut a dirty great hole in my lovely carbon bonnet!

Share this post


Link to post
Share on other sites
brumster

Well, the bonnet issue is not ideal but it does close up now - the top of the filter is touching the underside of the bonnet, but it'll suffice for now until I can get a reangled inlet manifold from Sandy.

 

Big ballache is that I need to ditch the alloy sump because it's just way to big, easily an inch lower so there is no chance of getting my sump guard on. So I'll need to swap the spacer and tin sump on from my alloy 1.9 - along with the PTS baffle kit, of course. Ho hum. Next event is a run around Rockingham as a bit of a shakedown, and the good thing is it's so billiard-table smooth I don't bother running a sump guard there, so there is no rush to do the swap thankfully.

 

Will spend a few hours tomorrow mapping it up on the road, keen to see how it's going to feel.

Edited by brumster

Share this post


Link to post
Share on other sites
Guest 205dangti

could u solve the problem with an auto bonnet with the raised centre section. if so ive got one for sale and im in birmingham. 60 quid if u can collect. got a few dents but rare now.

Share this post


Link to post
Share on other sites
welshpug

doubtful, as the clearance is required at the front not the middle/back.

Share this post


Link to post
Share on other sites
brumster

No, as Welshpug says, the auto bonnet doesn't really have the bump in the useful place. I had one before but it didn't really make any difference. Thanks for the offer, though.

Share this post


Link to post
Share on other sites
brumster

Well, the weekend just re-enforced the fact that nothing goes as smooth as you hope :(

 

Went to put the strut brace in and discovered the plastic cowling over the cam pulleys fouls the bar, so had to take a hacksaw to that :( (the cover, not the strut brace!)

 

The wideband lambda sensor I've got here fouls the gear linkage when inserted into the boss, such that I can't get 1st and 2nd, so I didn't get round to mapping it on Sunday after all. Long term option I'm going to need to replace a section of the system and weld a new boss in at a slightly different position. Thankfully the existing map is enough to potter around up to 3k (at which point it goes a bit flat), so I came up with a backup option to drive out to a nice dual carriageway with two large roundabouts at each end, pull over, swap in the sensor then get moving in 3rd and provided I keep moving, I should be able to do the majority of the mapping just whizzing up and down in 3/4/5th. Bit of a PITA though, so will see how quickly I can get the boss moved for the lambda (or find an alternative sensor where the body isn't so long and fouls the linkage).

 

Took it for a spin around the block anyway, just to make sure everything held up, and it's fine. The 6-paddle solid clutch is a little 'different', not heavy but just clattery to pull away with... just something to get used to I guess. Doesn't seem overly different to the old engine yet, but then again with no proper fuelling and a 3k useable rev range this is hardly surprising... need to put a few miles on and then unleash the other 5500 :)

 

Dragging on longer than I'd hoped for. Knock on effects like the sump needing changing, the oil cooler pipes lengthening and so forth have meant it wasn't as simple a swap as I'd have hoped.

Edited by brumster

Share this post


Link to post
Share on other sites
Daz_C

Rather than replace a section of the system why don't you just plug the existing boss and fit a new one in a more convenient place ?

Share this post


Link to post
Share on other sites
brumster

Yes, very true :rolleyes:

 

Got the lambda issue sorted out last night so went out for an hour and did a bit of mapping. The self-learn facility, if you don't know how these things work, trims the fuel map according to a target air/fuel ratio that you set in the ECU for the various load sites and revs breakpoints. This means you effectively stick the lambda in, turn the self-learn on, and just 'drive around', trying to hit all the various combinations of throttle openings at different load sites that you can. The ECU has various parameters that control how much, and how when, it adjusts the fuelling and it basically builds up a learn map - a multiplication table that adjusts the main fuel map up to a factor of 2.

 

We put in a reconfigured advance map first, and then you just drive around and it pretty much sorts itself out. At the end, you merge the learn map into the main fuel map, therefore making the changes permanent so you can clear the learn map, and redo it all over again. Repeat it a few times and it refines itself. Unfortunately I couldn't get the merge feature working in the software (the instructions are a little curious, shall we say) but that's something I can do later when OBR let me know what I'm doing wrong!

 

Next job is to smooth out the curves a little, there's a curious dip in the fueling around 4250rpm which throws it a coughing and spluttering fit, so that needs to be brought up - it's just ironing out the bugs now.

 

In terms of the swapover, at least I can offer some thoughts now.

 

How does the extra weight feel?

 

Well, honestly, I can't tell bugger all difference. We were fair pelting it around some local (quiet) country roads and it felt as sure footed as it ever did - bloody twitchy on the power over uneven surfaces, but it was like that before :wacko:. So I would say *if* you've got some semi-decent suspension on there already (coilover billies, adjustable arms, eccentric top mounts) then you won't need to be concerned about the extra 16Kg.

 

As for the engine, well, I must admit it didn't feel overwhelmingly more powerful until I asked my colleague if he thought it felt any quicker than the old one - he gave me a semi-curious look while he caught his breath, as if to suggest I was a bit mad, so he reckons yes it is :lol:. I then realised I'm only using 6 of the 8.5 grand of revs I've got to play with, so maybe there's more once it's all run in (the engine has done, like, 30 miles - I know "give it death" is a coined phrase but I don't actually want to kill it!).

 

The proof of the pudding will be Rockingham where it'll get some proper ragging and I should be able to guage how quicker it is but at the moment I guess it's a little 'tight' and needs to free up a little. Maybe behind the wheel you don't feel the additional grunt as you're too busy concentrating on keeping the thing straight!

Share this post


Link to post
Share on other sites
brumster

Well after putting a few more miles on it, I finished the mapping over the weekend. I was also a little more confident in 'giving it the beans' because I basically needed to in order to finish the mapping, so it was time to get it up there into the 8's and see how it goes.

 

I spent some offline time smoothing the map out in EFI's software; it's not the prettiest software I've seen but it is functional and one thing I like about it is it uses lots of keyboard shortcuts, so no reliance on having to use a mouse - clearly designed by a mapper, rather than an IT nerd :lol:. I also lifted up the fueling in the midrange which was a byproduct of my previous engine, which Richard told me was quite typical for a standard mi16 head - they liked leaning out in the midrange - but this new engine doesn't need it, presumably because it breathes better on the worked head. Uploaded and on the road, the fluffing at ~4200 has disappeared so I just ran up and down the dual carriageway at various load sites/revs, letting the self-learn on the wideband lambda do it's thing.

 

Piece of piss, really. Two merges of the learn map later, and it's running tickety-boo. Not massively urgent below 5000rpm (relatively speaking) but then it distinctly comes on-cam at 5 and positively roars up to my 8200 limit. It was actually a little scary, you can really feel the diff working hard and it tends to weave a little on hard power, although to be honest my front geometry really needs a good look-at, it's not had some attention for a long time and is probably off.

 

It's not Andy Baker quick, but it's certainly quick enough for me. I remember the jump from 1.9 8v to my mildly-tuned 1.9 mi16 felt light night and day. The jump up to big-league numbers hasn't felt like quite such a leap, but then the engine is still a bit 'tight' and maybe it'll free up over time. I think more likely it's the law of diminishing returns, coupled with just getting used to things.

 

She puts the power down okay though, mind you I'm on Dunlop X08's, but at least my concern that she'd spin away all that new-found grunt doesn't look like it'll be the case (well, not in the dry anyway). I think if Rockingham is like it was last year, they'll be little point in the extra 50 ponies B)

 

So engine up and running. Next updates are waiting on parts :-

 

1) Mocal pipes to extend oil cooler

2) Swap sump for tin type + baffle - will leave this until first oil change, put a few more miles in on the s*it oil first.

3) Long-term I'll probably get a new inlet manifold off Sandy, but funds-wise it'll have to wait until next year - it'll get by for now.

Share this post


Link to post
Share on other sites
Craigb

Nice one , Hope you don't need the outboard for rockingham

Share this post


Link to post
Share on other sites
brumster

Last of the little post-conversion jobs finished off at the weekend.

 

Tin sump

I whipped the old tin sump and spacer off the alloy block to fit on the S16 over the weekend - the alloy sump is too deep and simply doesn't allow my sump guard to fit, and it's too low to run without one. I needed to drain the run-in oil and re-fit the oil cooler anyway so all the jobs tied in with each other. Off came the old baffled XU10 sump, then the oil pump itself so I could get to the XU10 windage tray bolted to the main bearing cap bolts. It all came out, then the old spacer went into place. The two strengthening braces across the middle needed gringing off - not something you'd do on the original alloy block as I believe it's not just a spacer but also adds bracing to the block - but on the iron block I figure this bracing won't be needed (it won't fit anyway).

 

First problem was the PTS-pattern sump baffle. I am guessing Longman fitted a gti6 pump when building the engine, because dimensionally it's just slightly different in places and there was no way I could get the pump and the baffle to clear each other when fitting. I noticed that careful removal of a couple of edges of the vertical plates on the baffle would make it slide over the pump a LOT easier, without what would look like negligible negative affect on the effectiveness of the baffle itself. Out with the plasma cutter and presto, one much-easier fitted baffle kit. Whizzed it back on with the chain/sprocket guard too and bolted it all up, job done.

 

Also fitted some new lower arm and front ARB bushes in order to tame the scary wandering on power - these improved things a little but it was still the wrong side of worrying, but there was a worry amount of toe-in and a few quick turns of the steering arms and it was improved further. I could probably do with a little more fine-tuning/fiddling but it's getting to a manageable state now. Need to find a point where it's how I like it...

 

I think as far as the engine swap goes then, that's the story over, although I still haven't really been able to make a call on how it handles out there on the stages with the bigger engine up front. I'll post something after Rockingham that finally answers the question, then close the thread ;)

Share this post


Link to post
Share on other sites
brumster

Ok, I'm convinced now, that engine is a cracking engine!

 

I guess it's just the effortless torque it now has compared to the old one; the pull out of corners is definately much better now.

 

In summary :-

 

1) No oil surge issues

2) Plenty to suggest the 250 horses Longman spec is about right, seeing how we compared with some of the 250+ Mk.II monsters out there

3) Weight of block no difference with proper suspension; in fact might even have improved things a little (better grip)

4) Everyone was saying how nice it sounds from outside but from inside I can't verify this B) !!

 

It is about the limit for what you can realistically use out there, though, I could chase more horses with stupid specs but the rest of the car can't keep up now so I think this engine will probably be about the limit. Reiger suspension next in search of some grip and adjustability....

 

<stamp> Case closed :blush:

Share this post


Link to post
Share on other sites
andychalmers

My 205 rally car has a S16 engine. Standard engine on throttle bodies, producing about 165bhp, its not masses of power but has loads of grunt. The weight of the engine does'nt seem to make any difference. The cars fine in the forrest as well as tarmac. I run Bilstein suspension, 4.8cwp, close ratio box plus a tran-x lsd. Next step was a new S16 monster engine like yours but decided to build a super 1400 Swift rally car, should see 190-200bhp (its costing me enough).

Share this post


Link to post
Share on other sites
James_m

How did you go about sealing the hole in the side of the head when using the XU9J4 thermostat housing?

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
Sign in to follow this  

×