Phoenix from the flames - Porsche Boxster with an Audi 2.7T

Phoenix from the flames - Porsche Boxster with an Audi 2.7T

Author
Discussion

The Wookie

13,972 posts

229 months

Friday 25th August 2023
quotequote all
Not a happy pressure plate. Looks like contamination to me, doesn't even look like it's been that hot. The rest of the gunk around seems to correspond too.

Have you suffered an RMS or gearbox input shaft seal leak at any point? Alternatively what condition was the thrust bearing in, was the grease still contained within it?

Escy

Original Poster:

3,951 posts

150 months

Friday 25th August 2023
quotequote all
I don't think there's oil contamination, lots of clutch dust. There's no oil in the gearbox bell housing. When I've had oil leaks bad enough to cause a slipping clutch it's always been pretty obvious. The grease is in the thrust bearing.

Good news I think on a clutch. I saw Sachs Performance do a kit which consists of a lightweight single mass flywheel, the pressue plate I bought and a sprung organic plate. Rather than the solid plate they typically sell with their pressure plate. I wanted a sprung plate so hopefully this is me sorted. Darkside Developments have them in stock.

I've ordered it. Although it's not rated for the torque, the LUK clutch lasted pretty well so I can't see this having any issues. Just to be safe I think I'll take a bit of timing out of it as it comes on boost, it's spinning the wheels in 3rd gear anyway

Fastdruid

8,663 posts

153 months

Saturday 2nd September 2023
quotequote all
Escy said:
Good news I think on a clutch. I saw Sachs Performance do a kit which consists of a lightweight single mass flywheel, the pressue plate I bought and a sprung organic plate. Rather than the solid plate they typically sell with their pressure plate. I wanted a sprung plate so hopefully this is me sorted. Darkside Developments have them in stock.
I'd be interested to know the part number/link for that one. I have a 1UZ matched to a 01E gearbox with a mix of custom flywheel and Sachs pressure plate but I'm struggling to find a suitable sprung plate to fit as the springs on the RS4 sprung plate interfere with the Sachs plate.




Escy

Original Poster:

3,951 posts

150 months

Saturday 2nd September 2023
quotequote all
Fastdruid said:
I'd be interested to know the part number/link for that one. I have a 1UZ matched to a 01E gearbox with a mix of custom flywheel and Sachs pressure plate but I'm struggling to find a suitable sprung plate to fit as the springs on the RS4 sprung plate interfere with the Sachs plate.
I've been having a nightmare with this, I ordered a pressure plate that was for a Sachs kit for an RS4, I didn't take into account it's designed for a solid disc to be fitted to a dual mass flywheel. This is the disc I bought to match it fouled the pressure plate, only slightly but I wasn't willing to risk trimming stuff then find out the disc is too thick or something and then it's all scrap. I took the 25% re-stocking hit on it and sent it all back.

https://clutchstop.co.uk/products/sachs-performanc...

It's been such a frustrating experience trying to find something. Aftermarket stuff is poorly served in the UK, everyone proudly advertises stuff but nobody keeps stock of anything. I spoke to CG Motorsport about their twin disc clutch, 3 weeks to make one. I could of ordered another Sachs pressure plate (should have gone for the one out of their kit they do with a single mass flywheel) but it's a punt that it'll be right and 3 weeks to wait to find out. Tegiwa offer ACT clutches, again 3 weeks wait. I looked at replacing my B7 RS4 flywheel with a B5 RS4 one and getting a clutch kit for that, there are a few reasonably priced (but maybe not good) options like Drive Torque and Black Diamond, spoke to retailers and again no stock and 3-4 weeks.

I've ended up buying a clutch from Ringer Racing in the US. I'd initially discounted their stuff on price but I'm sick of messing about. They are in stock, and it was a 3 day delivery all the way from America. It arrives Monday. They are highly rated on the Facebook groups for the power they hold and the way they feel so I'm confident I'll be happy with it.


Fastdruid

8,663 posts

153 months

Saturday 2nd September 2023
quotequote all
Escy said:
Fastdruid said:
I'd be interested to know the part number/link for that one. I have a 1UZ matched to a 01E gearbox with a mix of custom flywheel and Sachs pressure plate but I'm struggling to find a suitable sprung plate to fit as the springs on the RS4 sprung plate interfere with the Sachs plate.
I've been having a nightmare with this, I ordered a pressure plate that was for a Sachs kit for an RS4, I didn't take into account it's designed for a solid disc to be fitted to a dual mass flywheel. This is the disc I bought to match it fouled the pressure plate, only slightly but I wasn't willing to risk trimming stuff then find out the disc is too thick or something and then it's all scrap. I took the 25% re-stocking hit on it and sent it all back.
I feel your pain. So many places that use "stock" photos, don't give critical dimensions or details (and I include Sachs, LuK etc themselves here). I've currently got a disc that fits (but isn't sprung) and a disc that is sprung that fouls. Annoyingly it fouls on a non-critical part of the pressure plate too. I've considered milling it off but in a similar way to you I'm loath to start modifying in case it doesn't work and then I've got a ruined pressure plate and still don't have a solution.


Escy

Original Poster:

3,951 posts

150 months

Wednesday 13th September 2023
quotequote all
Generic photos are a pain in the arse when it comes to stuff like this.

The Ringer Racing clutch arrived in 3 days from America. It's a stage 3 rated for 715ft/lbs so should be plenty. The disc is organic one side, Cerametalic the other side.

IMG_20230904_173436764

IMG_20230904_173425555

IMG_20230904_201114119

The clutch slave cylinder had shown signs of a slight leak, my other car broke down so I had to just get it back together as it was. I'm still bedding the clutch in but first impressions are good, the pedal is slightly heavier but would still pass as OEM. There's a little bit of soft juddering if you let it engage without enough rev's. The biting point isn't that obvious which is good, it's not grabby at all. It's definately not ruined the car so was worth the money, I'm happy with it (providing it holds the torque).

I had to take it all back apart a few days later to replace the slave cylinder. As I needed to bleed it, I took the opportunity to upgrade the brake fluid to Dot 5.1 brake fluid.

IMG_20230910_154500101_HDR

IMG_20230910_154507511_HDR

I had some help on the pedals.

IMG_20230909_172257979

There was an oil leak on the cam cover, it's the Achilles heel of these engines. Such a piss poor design, 3 bolts per side and 2 in the middle, It's just not enough. Most cam covers would have double that amount.

IMG_20230908_191518076_HDR

Fastdruid

8,663 posts

153 months

Wednesday 13th September 2023
quotequote all
Escy said:
Generic photos are a pain in the arse when it comes to stuff like this.

The Ringer Racing clutch arrived in 3 days from America. It's a stage 3 rated for 715ft/lbs so should be plenty. The disc is organic one side, Cerametalic the other side.

IMG_20230904_173436764

IMG_20230904_173425555
Interesting. That's a relined LUK clutch disc. This is a photo of an original 324 0371 10 LUK clutch disc which is organic both sides (the one on the left to save anyone peering at the numbers!)



Same sprung centre though as the one I have which means it still wouldn't work for mine (mine is an unholy mix of a flywheel to mate a BMW gearbox to a Lexus/Toyota engine and an Audi gearbox).

Escy

Original Poster:

3,951 posts

150 months

Wednesday 27th September 2023
quotequote all
I bought a load of pipes to take care of my coolant leak.

IMG_20230910_150027071_HDR

It didn't go to plan I couldn't get some of the parts apart and needed to buy more. I tried my best to get them apart, I made a tool so I could attach a slide hammer to the pipe. It just wasn't having it.

IMG_20230911_204231660_HDR

So I had to buy some more bits. Porsche Cardiff said the big plastic pipe was on back order and it was looking like 3 weeks, luckily Porsche Swindon had everything I needed in stock and they sent them out next day delivery

IMG_20230913_181503596

No wonder I couldn't get them apart.

IMG_20230911_232541645_HDR

IMG_20230911_232218235

It ended up being a bigger job than I'd expected. I'm glad it's done as it's one of the common 9x7 issues.

IMG_20230913_211857104

As the subframe was off I needed to do a wheel alignment. I've bought a new wheel alignment system, made by Vamag. I like it, it's a good system, it's newer than the other ones I had and it seems like it's had less use. It also has some upgraded wheel clamps which are better than all the other ones I've used. It's on Windows 10 which means I can easily share the screen with an android tablet which makes it much easier to see what I'm doing when I'm in the pit. I've added a bit more caster on (was around 6.5 degrees before), it's made so much difference to the way it drives, the steering feels sharper.

IMG_20230916_115823005

IMG_20230922_161302040_HDR

I've been out for a blast in it, trying to solve an over boost issue. I set the boost control up during the hot bit of the summer and I think the colder ambient temperatures have made a difference. That means my PID control loop for the boost solenoid wasn't set up correctly. Too much of the heavy lifting was done by the duty cycle table which the PID control uses as a baseline. If the duty cycle is bang on the money the PID control loop isn't needing to do that much work to trim it. I don't want to be manually changing the duty cycle all the time, I want to PID to work over a larger range to be able to compensate and keep the boost on it's target.

I don't really know how to dial in a PID (and if you actually ask a professional tuner about it, I don't think all that many do either). It's been a case of change something, make a pull and see how it is. I did about 20 pulls and ended up no better off than when I started. The main issue is I've got an initial boost spike then a dip before it flattens out. You can feel this on the road. The good news is the new clutch held up fine and it was good fun. It's first time I've been able to drive it hard since July.

I am also going to need to turn the boost down in 3rd gear, it spins the wheels every time it comes on boost.

Jester86

438 posts

110 months

Wednesday 27th September 2023
quotequote all
As always, love reading the updates!

Dr G

15,212 posts

243 months

Wednesday 27th September 2023
quotequote all
Same, good progress smile

Your light judder (with no revs) sounds pretty normal for an aftermarket/uprated type clutch. Sachs Organic in mine does similar at crawling speeds but nothing with a few revs on.

The Wookie

13,972 posts

229 months

Thursday 28th September 2023
quotequote all
Escy said:
I don't really know how to dial in a PID (and if you actually ask a professional tuner about it, I don't think all that many do either). It's been a case of change something, make a pull and see how it is. I did about 20 pulls and ended up no better off than when I started. The main issue is I've got an initial boost spike then a dip before it flattens out. You can feel this on the road. The good news is the new clutch held up fine and it was good fun. It's first time I've been able to drive it hard since July.
No experience of boost control but a little bit with PID on hydraulic pressure control systems. Might be worth having a read of this:

https://www.marpledata.com/blog/how-to-tune-a-pid-...

I bet if you delve into the details of the ECU you'll find all sort of parameters for bounding the integrator.

You may have to sacrifice a bit of response to get a stable system with no overshoot as ultimately you're limited by the sheer physics of the system in terms of the speed of actuator vs the time for the turbo to speed up/air to change velocity/exhaust pressure to increase/sensor to measure the change/etc etc etc.

Edited by The Wookie on Thursday 28th September 09:35

Richyvrlimited

1,826 posts

164 months

Thursday 28th September 2023
quotequote all
Escy said:
I don't really know how to dial in a PID (and if you actually ask a professional tuner about it, I don't think all that many do either). It's been a case of change something, make a pull and see how it is. I did about 20 pulls and ended up no better off than when I started. The main issue is I've got an initial boost spike then a dip before it flattens out. You can feel this on the road. The good news is the new clutch held up fine and it was good fun. It's first time I've been able to drive it hard since July.

I am by far an expert, but tuning PID on a Megasquirt ECU is done by:

Adjust the I value to get you to the target
Now adjust the P term in changing conditions to keep you at the target
D is used as a dampener, so may not be needed, basically leave alone unelss you specifically need to touch it.

it sounds like you may have too much D so it's taking a while to pull you back down to target after overboosting and maybe too much P as you have an oscillation.

Escy

Original Poster:

3,951 posts

150 months

Thursday 28th September 2023
quotequote all
I thought I'd post some screenshots to show what's going on.

This is a log from July pretty much the last time it was driven hard before the clutch went. It's a pull from low revs to the redline in 4th gear. You can see how tight the boost follows the target. I was happy with that but like I said before, it was mostly achieved by calibrating the duty cycle table. Now the air is more dense the same duty cycle percentage makes more boost and it's putting it past the control of the PID control loop.



This is the boost delivery at the moment. It overshoots the target, then dips under before settling on it. If I change gear it'll spike again when I come back on the throttle (which is one of the reasons I like the flat foot shifting as the boost doesn't spike).




The Wookie said:
No experience of boost control but a little bit with PID on hydraulic pressure control systems. Might be worth having a read of this:

https://www.marpledata.com/blog/how-to-tune-a-pid-...

I bet if you delve into the details of the ECU you'll find all sort of parameters for bounding the integrator.

You may have to sacrifice a bit of response to get a stable system with no overshoot as ultimately you're limited by the sheer physics of the system in terms of the speed of actuator vs the time for the turbo to speed up/air to change velocity/exhaust pressure to increase/sensor to measure the change/etc etc etc.
Thanks for the link. You might be right, I've got my boost duty % set high at the low end of the table where it can't make the boost and then drop it down as it comes on boost to improve the response. Also my P and I values are different based on how far I am from the target. I might be trying to do too much there. I'm probably better off getting simplifying it all and then try to add in some improvements in response afterwards.

Richyvrlimited said:
I am by far an expert, but tuning PID on a Megasquirt ECU is done by:

Adjust the I value to get you to the target
Now adjust the P term in changing conditions to keep you at the target
D is used as a dampener, so may not be needed, basically leave alone unelss you specifically need to touch it.

it sounds like you may have too much D so it's taking a while to pull you back down to target after overboosting and maybe too much P as you have an oscillation.
The link Wookie posted said similar regarding the D. I'm going to zero the gain as a starting point.

Escy

Original Poster:

3,951 posts

150 months

Sunday 8th October 2023
quotequote all
It might be the last decent weekend before the winter so rather than waste it doing DIY (my wife didn't share this sentiment) I went out for a drive around the mountains with a few of the guys on the South Wales Porsche Facebook group. I've not actually used the car that much this year so it was great to get out for a rip. The roads and views are amazing but we didn't stop anywhere picturesque for photos so this is all I got.

IMG_20231007_162038534

IMG_20231007_135057110

It was 2 years ago the last time I went on a road trip like this and while I loved it, it proved hard on the car which kind of fell apart. There were ECU problems that kept limiting the engine power, the suspension was a bit soft, a remote reservoir on my coilovers hit the body and ended up dragging along on the road and my ABS was kicking in all the time and ruining things. I also reversed into a rock and messed the bumper up (I can't blame the car for that one). These sort of drives really test out a car.

This time around the car performed really well, I ran it at full boost the whole time, it got a total spanking and it was awesome. It feels so fast on roads like that, such an adrenaline rush. All the development over the last 2 years kind of gets lost but this drive proved the difference. The brakes are sorted, the ABS no longer intrudes like before. Handling wise the car is on point, it coped with all the dips, bumps and undulations really well. The boost pipe issue is sorted now, I've still got no oil leaks. The ECU had a firmware update and the knock control works now so that allows me to relax a lot more and just enjoy the driving. I finished the road trip feeling really happy with it all, no list of work to do this time.

There was a 991 Turbo PDK there, the owner said it was 600bhp. At one point he led the way and I was behind, I could match it for pace so my Boxster is doing alright.

Then on the way back I had a bit of fun with a motorbike and it had a misfire on cylinder 1 at full throttle. It's the green line you can see that's bellow all the others. A little disappointing after how good I was feeling about the car.

cylinderdown

When I got back home I swapped the spark plug from cylinder 1 to 3 and the coilpack from 1 to 6 and took it back out for another drive. No issues at all, ended up doing 20+ pulls while tinkering with the boost control and it never happened again. Maybe I'll chuck a fresh set of spark plugs in it.

A couple of pictures I took on the way home.

PA071658

PA071660

PA071662

PA071661

HughG

3,550 posts

242 months

Sunday 8th October 2023
quotequote all
Wonderful, it’s a credit to your perseverance that it has got to this stage, I’m glad you enjoyed it!

shalmaneser

5,936 posts

196 months

Sunday 8th October 2023
quotequote all
Such a credit to your perseverance. Awesome car and must go like a stabbed rat.

dobbo_

14,402 posts

249 months

Sunday 8th October 2023
quotequote all
I love this thread, I am not a fan of Boxters but yours is ace!

Escy

Original Poster:

3,951 posts

150 months

Monday 16th October 2023
quotequote all
HughG said:
Wonderful, it’s a credit to your perseverance that it has got to this stage, I’m glad you enjoyed it!
It's starting to feel like it's paying off now. I'm really happy with how it all is.


dobbo_ said:
I love this thread, I am not a fan of Boxters but yours is ace!
Thank you.

shalmaneser said:
Such a credit to your perseverance. Awesome car and must go like a stabbed rat.
It's going pretty well at the moment. I took it out this evening, did a little tinkering with the tune and I raised the rev limit up 200rpm to 7500rpm so it'll do 200kph in 4th gear. The idea being to have a crack at some 100-200kph times on my mates private road. My previous best was 6.97 which I knew I could improve. The best I managed this time was a 6.22 which I am pretty happy with. That wasn't a clean run either, there were 2 traction control ignition cuts, the second one at 92mph. It's never spun the wheels in 4th gear before, that's pretty spicy!

That time can be improved upon, I'll take some ignition timing out on the midrange to stop the wheel spin. I was comparing it to stuff on the Dragy app, it's in and around the times done by a Huracan Performante or a Ferrari 488. Not bad for a Boxster.

Screenshot_20231015-233254

Escy

Original Poster:

3,951 posts

150 months

Saturday 2nd December 2023
quotequote all
A bit of an update, I've barely used it the last couple of months but that's not stopped it being a pain in the arse.

There's been a lot of rain, I noticed loads of condensation on the inside of the windscreen. I had a feel under the carpet behind the passenger seat and it's soaking wet. You can't really see in the photo but there was a puddle in the footwell.

IMG_20231105_145726622

Usually the first time you know you have a water leak is when you get strange electrical happenings when these modules under the seat get wet. Luckily they weren't. I've wrapped them in plastic bags, they aren't sealed but still a bit more protection. I should have done this last time but naively thought once I've fixed that leak I'd be trouble free.

IMG_20231105_145733705_HDR

I dried it all out and got someone to run a hose pipe over the rear end while I looked for leaks. There was nothing. I then realised the leak was actually coming from the front.

IMG_20231117_120820189

I stripped the whole interior so I could remove the carpet. The carpet is basically a huge sponge, It's surprising how much water it soaks up. The carpet must have weighed about 25kg so there's got to be 10-15 litres of water in it. I've left it 3 weeks and it's not dried. I still have the original carpet which was swapped out a year ago the last time the car leaked so that is going back in.

IMG_20231127_205443088 by Thomas Prosser, on Flickr

It's leaking from the front behind this black panel. There are drains in the scuttle area but my car doesn't pool water there and when I pour some in it runs straight out. It also runs straight into the cabin which rules out an issue with the pollen filter or the windscreen which can be another common cause (my windscreen is original).

IMG_20231118_155856544

This is the panel from the other side.


IMG_20231118_160026268

Here's what it looks like. You can see the seal on it clearly in the second photo. It's the same rubberised foam when use on the door panels which also fall apart.

99750412101_LR3

99750412101_LR4

It looks like when Porsche built these cars they started with this panel as the first piece. It'll be so much work to replace it. It would involve removing the A/C gas, taking the dashboard and HVAC unit out. I'll need the whole front end stripped down so I can get the wiring looms to pass through it. I'll need to disconnect the clutch hydraulics (which means taking the exhaust off when I bleed it) and possibly brake hydraulics. There's probably more stuff I'm over looking. I couldn't see any mention of this panel leaking when looking around Facebook and forums. I spoke to a local Porsche specialist who told me he's seen a few cars with this leak. He charges 2 days of labour, so would be around £1500l. The part is £200 plus the cost of an A/C re-gas. Obviously I'd do it all myself so it wouldn't cost me much but I don't really fancy it.

Someone told me they had a leak here and they sealed it with Captain Tolley's which is a creeping crack cure mainly used for boats. I figured I'd give it a go. It's supposed to be used at over 10 degrees, it's been colder than that.

Screenshot_20231202-130623

Applying it has been a challenge, I've been using a pipe in the end of the bottle and a bore scope camera to see what I'm doing as I cant get my fingers anywhere near it. Over the period of a week, I'd apply it, let it cure (but it's too cold) then when I'd test it (with a hose pipe directly in the bulkhead), twice I'd ended up with a very small puddle on the footwell which was milky white (this stuff is milky white but cures clear). A hosepipe in the bulkhead might have been a bit vigorous. I've now applied a load more but not tested it (other than an air line from inside the car when it was all wet, to see I couldn't see the fluid moving). The plan is keep it dry for as long as possible and see what happens next time it's raining.

Screenshot_20231202-130547

In retrospect have probably been better off just buying the panel and getting stuck in, I still don't know if it's fixed and I could have done it properly by now. It was removing and getting the A/C gas put back in that put me off. The A/C was gassed up 2 years ago and it's never worked. It's a bit complicated as you need to be sending and receiving canbus messages. I assumed this was my issue and went down a rabbit hole trying to work it all out but it's also been on the back burner as it's a convertible so no A/C never became a massive issue.

I hadn't paid enough attention to the A/C compressor when I bought it. I got a second hand one off ebay. At the time there wasn't a 2.7T one so I got one off a newer Audi A4 B6 3.0 V6. I was more concerned about the mounting bracket and if it would physically fit my engine. It turns out that this compressor didn't have a clutch like the original Boxster or Audi 2.7T compressor did. The pulley is constantly driven and it uses a variable solenoid to control the flow of gas. Giving it 12v I'd get a click to indicate it was opening so I knew I was ok there. I'd tried running it off a PWM directly from the ECU. I also wrote a control strategy on my ECU based off pressure. It just wouldn't work.

After doing all the complicated stuff and double checking it all it turns out the pulley on the compressor was broken. I put a spanner on the end and it turned easily, it shouldn't with the belt on. The rubber dampener inside had been worn away. I didn't want to take the compressor off due to the gas. The only place I could find a new pulley was AliExpress.

IMG_20231128_183950142_HDR

This was straight forward to fit. Another job that's fair easier on my car than it would be on an Audi S4 or a standard Boxster.

IMG_20231128_190334839_HDR

The next day I needed to use the car so for something important so I didn't get a chance to test it. I did notice the idle would drop low and occasionally stall. Once I got back home I parked in the garage forwards and tested out the A/C. It started making some dodgy noises, bangs clanks and clatters. I turned off the A/C it still made dodgy noises, I didn't know what it was but I wanted to get the car out of the garage and turned around so I could get it over the pit before something packed up. I was under duress, the engine kept cutting out. I ended up kerbing my left rear wheel twice (on the same kerb I damaged the old bumper and broke a light on 2 years ago). Think I've damaged every wheel now.

IMG_20231202_124828635_HDR

When looking under the car, the front engine mount I had made had snapped. I didn't take a good photo but you can see where it broke. It looks like the weld has failed half way around and it's snapped steel on the other side. It was done years ago and it's done plenty of miles since. I'd looked at the engine mount the day before when doing the A/C pulley and it was fine.

IMG_20231129_190856982_HDR

It would be a strange coincidence so I'm convinced the A/C compressor has contributed to that failure. I made a new engine mount. Same design as before but using thicker steel and higher power setting on the welder.

IMG_20231201_143208525_HDR

IMG_20231202_112018731

Once the mount was sorted I could start up the engine again. I turned the A/C on and it made no noise at all. I checked the HVAC system on my diagnostics, there was no system pressure! The compressor won't do anything without pressure so that makes sense. I'm fairly sure it had pressure when it was making the noises as the idle would drop as the A/C was switched on and I was trying to tune that. I don't know where the gas has gone, I'd expect to see a wet spot where it's leaked out and I'd defiantly have heard and smelled it in the garage. All a bit odd, I'm confused by it. Not wanting to empty the A/C systems was the reason why I didn't just change the compressor to one with a clutch years ago and part of the reason I didn't want to commit to changing the bulkhead panel and now it's empty anyway.

Whilst I was working on it, I noticed a leak from my water/methanol solenoid. These push fit joints are utter trash, I don't know why the kit manufacturers use them, I seem to be working my way through the system replacing them one at a time as they fail.

IMG_20231202_124705114_HDR

IMG_20231202_124732112

I've managed to smash up the under trays on both sides. I think the majority of it happened when going on my platforms for the wheel alignment I did. I know how to avoid that happening next time but if they get mashed up again I'll need to raise the suspension a bit.

IMG_20231202_124902930

Considering the car has done about 20 miles in the last 2 months it's not exactly going well.

ReformedPistonhead

965 posts

138 months

Sunday 3rd December 2023
quotequote all
Amazing skills and perseverance well done