|
|
|
May 17, 2021 12:52:19 GMT
|
Nice work on the MOT. Hope you can get the small niggles sorted without too much drama. It is fustrating to fail on a slight leak at a caliper fitting though, alot of hassle for them to log it as a fail and have to retest vs the tester just giving it a quick nip up when he noticed it. As a tester, the problem with giving it a quick nip up is that if something breaks or you make it leak worse then It's down to us and not the presenter |
| |
|
|
|
|
gryphon
Club Retro Rides Member
Posts: 330 
Club RR Member Number: 157
|
|
May 17, 2021 13:19:14 GMT
|
|
The test station I go to is test only, no workshop and they very literally have no tools around. If I'd had my toolkit in the boot I could probably have nipped it up - either immediately, or on the ramp after the test and got an immediate re-test.
A bit annoying, but I'm not really complaining - they test fairly and we get on well.
|
| |
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
Jun 21, 2021 10:26:37 GMT
|
Once MOT'd the car was back in daily use for a bit - some top down commuting in the sun was lovely, even if it screams a bit on dual carriageway, which is half of my commute. The misfires became more noticeable, then the car started dropping out occasionally at idle, then very occasionally I'd be driving along and the engine would suddenly bog down instead of accelerating, then fix itself a few seconds later. I set up my phone to data log whenever the engine was running, and all of a sudden it behaved perfectly for around 400 miles! I have lots of data logged... Finally on my way home one day it bogged down on throttle, I've never been so happy for something to go wrong! I gave the throttle a couple of 100% blips to mark the failure time in the data. Exported the logs into Excel and started by graphing all the main sensors with RPM and throttle. The culprit became pretty obsious:  That's RPM, throttle and the Manifold Air Pressure. Notice how the MAP pretty much mimics the throttle, as you'd expect, then flatlines to a random value just before the throttle spikes? Found the cluprit! If it were a connector or harness failing, it would either jump high or low, and I can't work out how some physical obstruction in the vacuum line to it would cause that signal, so my bets were on the sensor itself. The sensor housing is a very commons shape, with multiple different part numbers having identical housings, but different keying on the electrical connector. The one supplied with the kit has no part number on it, and is totally unbranded, so is probably one of the many cheapy chinesey ones that pop up on ebay when you search for them. After some trial and error (ordered the wrong one first time round!) I worked out that the Ford part number for the sensor was 1144809 - although it's used by a multitude of OEMs with many different part numbers. A few decent suppliers made them, and I went with a Delphi one (PS10075).  All plugged in - I'm not 100% sure it's the right sensor still, not sure if the car will even start. Starts right up, better than it's started in a long time. Settles into a rock solid idle, none of the normal hunting. Pull away, definitely smoother, the revs used to dip a little more as the clutch engaged. Few miles of test drive - no more misfires and the engine feels a lot happier. I couldn't tell you exactly why, but it's definitely running better! I've not driven enough to prove out that it no longer bogs down occasionally and that it no longer dies every now and again at idle, but I'm optimistic! Finally getting a proper geo and cornerweighting done this afternoon too, so looking forward to that - and to seeing what the Rocketeer swap has done to the car's weights!  To finish with a picture of the car on commuting duties... I quite liked the small > modern small > large lineup. The Mazda 2 is a 'supermini' and so much bigger than the mx5, and is in turn dwarfed by the Evoque... or is it a Velar? |
| |
|
|
|
|
|
Jun 21, 2021 14:07:35 GMT
|
|
Good detective work on the issue and part number.
|
| |
|
|
jonomisfit
Club Retro Rides Member
Posts: 1,789
Club RR Member Number: 49
|
|
|
|
|
Modern car sizes are bonkers.
I've got a modern fiesta (b segment car?) and its taller and as long than my mid 90s escort which is c segment car. Its also about 200kg heavier.
When you have an original mx5 it really does bring back the perspective of how huge modern cars are.
|
| |
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
Jun 22, 2021 10:10:16 GMT
|
A page or so earlier in the thread I finally got the suspension to the point where it was worth getting a proper geo done - at which point I found out there was a bit of a waiting list! A month or so later it finally happened.  Check out that ramp-ception! Awesome bit of kit. (And I know, I know, lifting an MX5 on the sills... It's the only place you can lift it on this ramp, and I'm confident enough that my sills are solid) Prior to being brought in they asked that I set up the ride height to where I wanted it. Previously I'd popped the Meisterr coilovers on at their stock settings and it was a little low for my tastes. Looked good, but my numberplate and splitter gained a few scrapes from speed bumps, and I was very aware of my sump being a little lower than the subframe over bumps. The ride height adjustment ended up being a slightly rushed job over the weekend. I measured the front and rear heights of the other NA we have (wheel centre to wheelarch lip), which was around 10mm higher than mine at the front and 20mm higher at the rear. I went ahead and adjusted the coilovers on my car to match using the bottom cups to adjust the height rather than the spring seats... totally forgetting to account for the damper leverage. (1mm height change on the damper is >1mm at the hub due to the damper being attached closer to the lower arms pivot point) A little more faffing brought the car closer to where I wanted it, keeping the damper lengths equal across each axle, but when I got the car back onto flat tarmac for measuring, the rear left wheel had a far bigger arch gap than the rear right - by about 15mm! I have no idea where that came from. To fix it I shortened the offending damper by 10mm, again using the bottom cup. The ride heights looked better, but I now had unequal damper lengths across the rear axle. Fast forward to Monday afternoon - I arrive for the alignment and mention the ride height issue I had... I'm quickly informed that I've done it wrong, the dampers should be equal length across the axle and any changes in ride height across the axle should be compensated for with the spring preload. Oops! Matt gets it on the ramp and tweaks all the damper lengths, measuring from the top of the damper body to the top of the lower cup and brings them back to equal across each axle - with a fair bit more precision than I managed. Next, the bit I'd been waiting for - it goes on the scales  Annoyingly, as I bought the car as a non runner I never had it weighed before I started the conversion, but on the plus side, we have a very similarly specced, same year MX5 to compare against that had been weighed here a few years previously. So...  Well, the car is a little over a Ton, maybe a few Kg on the porky side for a late Mk1. Front axle is 549Kg / 53.3%, Rear axle is 480.5Kg / 46.7% Right hand side is overall 18.5kg heavier than the left - probably accounting for the battery, spare wheel, steering wheel, pedal box, brake/clutch systems and coolant header tank all being on the right - offset but just the power steering reservoir and washer bottle on the left. The good news was that immediately the cross car weights - FR+RL and FL+RR, the values that can be tuned by adjusting the suspension heights - were pretty much bang on at 49.9% to 51.1%. No need to tweak anything, meaning that the chassis is nice and straight, and the Meisterr springs are repeatable side to side. Next up was a geo on the hunter rig:   My old rear toe setting was pretty close, front not so much! My guessed by eye camber and castor were in the right sort of ballpark... ish. Matt pulled it into shape without issues, although the achievable front camber was a little lower than you can generally get on MX5's - potentially down to the new subframe, but hard to tell really. So - geo sorted. Let's take a look at Rocketeer VS Standard weights! Now bear in mind that this isn't a before/after comparison, they're two different cars with some differences between them. They both started life as '97 UK 1.8 Mazda MX5s. I looked at the VINs a while back and I think they were about 400 cars apart. The v6 has heavier brakes and discs (NBFL Sport Brakes vs Standard 1.8 brakes) but lighter wheels ( Oz Superlegerra vs Konig Hypergram), both on the same coilovers and bushes. Rocketeer has NB seats which I think are marginally heavier. The Dakar (standard MX5) had a style bar in the rear at the time which would add a few kg. Probably more differences that I've overlooked too - and I think the v6 had a touch less fuel in it. Rocketeer left - Dakar right:   So - first up, the Rocketeer is heavier by 21Kg. Slightly worse is that all this weight is added to the front axle, as the rear axle weights are identical! This brings the F/R balance to 53.3/46.7%, vs the 52.4/47.6% of the Dakar. I had vauge hopes of seeing some of the rumored 5Kg weight loss for the Rocketeer kit - but I think that is dry weight of engine/ancillaries/subframe and not the wet weight which might make some difference - although not 21Kg worth! As I said at the start though, we're comparing two different cars, who knows where mine started out. At the end of the day it's still a pretty light car by modern standards, it's now nicely balanced and geo'd with an excellent engine and I had a grin on my face on the way home! The steering response is now nicely symmetrical, it turns in a lot more confidently and holds it's line far more confidently. All in all you can lean on it far more than before and... I think I need to crack on with getting that Torsen diff back in there! |
| |
Last Edit: Jul 12, 2021 8:30:35 GMT by gryphon
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
Jun 22, 2021 10:33:44 GMT
|
Modern car sizes are bonkers. I've got a modern fiesta (b segment car?) and its taller and as long than my mid 90s escort which is c segment car. Its also about 200kg heavier. When you have an original mx5 it really does bring back the perspective of how huge modern cars are. Completely agreed - take a look at how big that ramp is in the last post compared to the car. He had to move the front wheel pads rearwards and inboard to make it fit too! There are some benefits to the added size and weight though. Every now and again I watch some Euro NCAP crash test videos and question my choice of commuting up and down the A46 in pre-NCAP cars... The one car that sits outside that trend in my mind is (predictably) the ND/Mk4 MX-5, which is smaller than my Mk1, weighs only a little more, and has a 4* NCAP rating. Pretty impressive - more of that please! |
| |
|
|
|
|
|
|
|
|
The corner weighting is very interesting! We did a bit of DIY corner weighting with my mate's turbo NA, my sister's 90hp 1.6 NA and my turbo X1/9. There's probably less accuracy in it than a professional one, but we're pretty confident we've got it pretty good. We even tried to even it all out to 3/4 of a fuel tank by calculating the weight of any missing fuel and adding it over the fuel tanks.
Headlines are that his turbo NA weighed in at 1019kg (including a relatively substantial rear roll bar). 51.9% front (additional weight of turbo offset slightly by the roll bar).
My sister's NA was a featherlight 930kg and 52.1% front, but it really is a very basic NA.
|
| |
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
|
|
The corner weighting is very interesting! We did a bit of DIY corner weighting with my mate's turbo NA, my sister's 90hp 1.6 NA and my turbo X1/9. There's probably less accuracy in it than a professional one, but we're pretty confident we've got it pretty good. We even tried to even it all out to 3/4 of a fuel tank by calculating the weight of any missing fuel and adding it over the fuel tanks. Headlines are that his turbo NA weighed in at 1019kg (including a relatively substantial rear roll bar). 51.9% front (additional weight of turbo offset slightly by the roll bar). My sister's NA was a featherlight 930kg and 52.1% front, but it really is a very basic NA. I've looked at setting up something for DIY corner weighting... did you use a couple of scales per wheel? Those are good weights! Were they earlier cars? I believe they got heavier over the years of production - my '97 has side impact bars in the doors and various bits of underbody stiffening that our '92 definitely doesn't have. My measurements above are with a full tank - I don't know why but that's what Spires asks for... I know 75% is the standard. I've mostly just been driving the car again, definitely enjoying the new geo! It has always had a bit of powetrain lash and clunk from the rear, but this was rapidly getting louder. I really thought something was about to fall off the car, but after going around and checking all the bolts twice everything seemed good. A little googleing later pointed me in the direction of diff bushes. (If you've read my BMW thread, I'm getting quite used to bushes!) This diff was a £30 'emergency' replacement for my old clunky Torsen, completely unknown provenance and mileage - it could be from anywhere between '94 and '04! The worn bushes are letting the diff twist under load and it's started knocking the metal arms against the mounting points. Diff bushes look an absolute pain to replace, but a common fix is to add void fillers - polybushes that sit around the original rubber bush, fill the voids (funnily enough) and can be fitted by just unbolting and lowering the diff in the car and sliding them into place.  Looks easy enough. Popped the back of the car on axle stands, removed 6 nuts and lowered the diff on the jack. It took a little wiggling the diff to get the bushes to drop below the mounting studs enough to slide the top little doughnut into place. I greased the bushes with red rubber grease, slid the top ones in, and sat the lower ones in the bottom supporting plates:  Then just 6 nuts back on and ready for a test drive. I should have done this ages ago - and I think I want to do it to the other MX5 too! Powertrain feels far tighter, no more lash and no more odd noises. It's also improved the clutch feel noticeably as it's no longer twisting the diff as it engages. Another step in the right direction |
| |
|
|
|
|
|
|
|
|
Thanks for the review, I think I need to do mine. I have IL Motorsport uprated bushes in my that were new maybe 5 years ago and haven't done many miles, it's not too clunky and noisy but anything to make it better is worth a shot. Also living on borrowed time with the diff but a good TorSen is getting pricey now
|
| |
Project Diary1975 Viva / 1988 T25 Camper / 1989 Mini / 1991 MX5 / 1992 Mini / 1994 Saab 9000 / 1997 Saab 9000 / 2008 Saab 9-5
|
|
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
|
|
I'm not sure if I mentioned it in the thread or not, but a few months ago the clutch engagement point got very low in the pedal stroke quite suddenly, then the next day wouldn't disengage at all... I gave it a bleed, got lots of black fluid out, and this bleed lasted for a long time - right up until last week when I started the car to go to work one morning and couldn't get a gear at all. As I replaced the slave cylinder with a new one along with the engine swap, my money is on the master cylinder, so out it comes...  The fluid that came out is rather black - especially as it had a flush a couple of months ago!  Circlip off and innards out - still pretty grungey.  I cleaned out the cylinder then with some 400 grit rolled around a crumpled tissue to give it some pressure on the cylinder walls. Difficult to get a good photo, but this was the result:  There is a decent amount of pitting where the inner seal sits at the start of it's stroke - I had a refurb kit on the side ready to be fitted, but this is too far gone! Genuine Mazda masters are a bit over £100 and out of stock everywhere and most others are £35 unbranded 'OEM Quality' jobbies. I found a pagid one for around £70 so that's in the post... The only other change I've made is to the fuel rail pressure - a little while ago on the facebook group one of the Rocketeer founders revealed that all their engine tuning and calibration is done at 3.5bar fuel rail pressure, but the kit instructions state 3 bar - as such, everyone is running at 3 bar! As the ECU runs closed loop fueling it compensates during general driving and it's not an issue. The only time there is no compensation is on engine start... And one of the issues I've always had with the engine is poor cold starts - it will chug along really roughly at around 500rpm for 10 seconds and die on any throttle input, then pick itself up and run beautifully. A quick play with the fuel pressure regulator brought the pressure up to 3.5 bar and the ECU adapted very quickly - I barely noticed running rich. What was far more noticeable is it now starts perfectly nearly every time, which is lovely. |
| |
|
|
|
|
|
Aug 25, 2021 11:31:35 GMT
|
Nice work on the cold-start fuel pressure fix! Hopefully your clutch master does the trick. I've looked at setting up something for DIY corner weighting... did you use a couple of scales per wheel? Those are good weights! Were they earlier cars? I believe they got heavier over the years of production - my '97 has side impact bars in the doors and various bits of underbody stiffening that our '92 definitely doesn't have. My measurements above are with a full tank - I don't know why but that's what Spires asks for... I know 75% is the standard. They're both late ones interestingly. The turbo's a '96 and the NA is a '97. They're both pretty basic versions of them though. Neither has aircon, power steering, or electric windows. Wouldn't be surprised if there are some bolts missing on the NA either, given the condition it was in when we got it! We used a set of 4 cheapy 300kg scales from eBay like these, but not sure we'd recommend it for cars that are any heavier than ours. They'd read a smidge over, but it was tricky to stop them overloading even with a 1000kg car. It was fine testing without a driver, but when we did the weights with my mate in the driver's seat the RH front just tipped over to being overloaded. We worked it out in the end by removing bits off the front right until it would read again, then weighing the bits, but it's surprising how much the side-to-side weight can differ. There was also a fair bit of shimming involved to get the scales level and even. Slight variations made quite big differences (in the 10s of kg!). It did work though, and for a tiny fraction of the price of proper corner weighting scales (which you'd still need to shim to get level). What I really want is a set of these: They look much easier! Unfortunately although they're more affordable and easier to level than proper corner scales, by the time you've added up 4 of them it's still nearly £1000! |
| |
Last Edit: Aug 25, 2021 11:32:17 GMT by biturbo228
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
Aug 25, 2021 12:04:05 GMT
|
|
Interesting, thanks! The instrumented go-jacks are a nice idea. Load cells seem to be getting cheaper too - Aliexpress seem to be selling up to 500Kg ones for £15, although I find that difficult to believe...
When i was looking to DIY it I came to the conclusion that it was going to be cheaper in the long run just to pay for the occasional corner weighting, with the bonus that I wouldn't have a load of corner weighting parts to store somewhere.
|
| |
|
|
|
|
|
Aug 25, 2021 12:51:58 GMT
|
|
Oh...that makes it a much more affordable proposition (provided I can sort out the electronics to read them). Really you'd want 4x 200kg load cells and then you could do pretty much any car you fancied.
Yeah that was the conclusion we came to as well. Less faff to get someone else to do it, although the price was fine for us. Although the weighting jacks would be useful when working out where to place various bits and pieces around a project car.
|
| |
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
Oct 25, 2021 21:18:50 GMT
|
The car sat clutchelss for far longer than intended. We got back from a couple of weeks in Scotland just as the fuel shortage was all kicking off, and I suddenly needed to get it back on the road - purely because it had most of a tank of fuel! A new Pagid master cylinder went in:
Fitting was painless, but bleeding took aaages. I have a little hand vacuum pump that in my head should be really good for sucking oil through from the reservoir and bleeding the system. Unfortunately it really doesn't seem to work like that and after trying for ages went back to the tried and true one person on the pedal, one on the valve, open, pump, close, release. It took a long time to get a decent bleed this way too, I suspect because you are trying to bleed straight down against gravity. Got there in the end though, car back on the road and it had fuel!
Next up was a way more overdue fix. Way back in early 2020 when I was preparing the car for it's first MOT in my ownership - and with the new engine - I realised the indicator stalk was a little tempremental. This was down to a large crack in the plastic assembly that the metal stalk mounts to. In my attempts to fix it I managed to make it a lot worse and ended up holding it back in place with cable ties 'temporarily'. It mostly worked, but it was quite easy to accidentally flash the headlights when indicating if the dipped beam was on.
I needed a new pair of stalks, but in the last couple of years of production Mazda changed the NA stalk design from straight ones to bent ones, which are not only different to the NB stalks, there are also multiple variants of the bent stalks over those two-or-so years of production. When I finally found what I thought was the correct set to buy, they were of course the wrong ones and didn't fit. They ended up on a shelf and I put up with slightly dodgy indicators up until very recently. Now that the evenings are drawing in again though, lights are getting pretty important again and I thought I'd give it another shot.
The stalk carrier on the replacement switches is the same as the broken one on the originals, but removing it involves a really awkward plastic clip - which I had managed to totally destroy while trying to unclip back in 2020.
4 very stiff, very brittle, very awkward clips. Last time I used heat, a 12mm socket pushed over them to compress them all at once, then tried to do the final bit with a screwdriver, which resulted in this:
This time I have a 3d printer at my disposal, so time for a specialised tool!
The outer ring sits over the outer ring of the clip to centre it, then has a ledge of just the right diameter to catch the chamfers on the clips and push them in - but this doesn't quite get them far enough, so the little piece in the centre is chamfered to push the clips slightly further inboard, and sized to push them out through the hole. I thought this would make my life nice and easy, but it still took many attempts, painful thumbs, lots of heat and lots of pressure before the stalk carrier finally slipped out.
And the old vs new carrier:
That done, reassembly was painless:
It's so nice to not have to baby the stalk any more! The niggles list is slowly shrinking....
|
| |
|
|
gryphon
Club Retro Rides Member
Posts: 330
Club RR Member Number: 157
|
|
May 12, 2022 14:38:22 GMT
|
Shortly after that last post the MX-5 went into hibernation for the winter - which is a fancy way of saying it sat on the drive turning green for 6 months, but at least it didn't get salty. Before getting it back on the road, the first job was to get rid of some of the greenery from the soft top. Sometime last summer I'd spent a long time on the roof with a Meguiars mohair cleaning and sealing kit, and to be honest I didn't get great results. As the stitching on this one is giving up the ghost anyway I figured I'd attempt a more aggressive clean, so out came the Vanish powder and a scrubbing brush.  With a bit of warm water, this worked incredibly well, although it took about 10 rinses with the hose before no more suds were appearing. Keeping in line with non-car products, I sealed it with spray on NikWax Tech Wash for waterproofing shoes and outdoor gear.  Really pleased with the results - it's far blacker and beadier than it was after the Meguiars treatment, let's see how it lasts.  I now just need to wash and polish the rest of the car.... A few weeks after this I decided to get it back on the road. Plugged the battery back in, turned the key, let it prime for a while, and it starts perfectly  Then the clutch pedal goes straight to the floor with zero effort.  Turns out I have no oil in the clutch reservoir and an oily patch on the drive. Attempting to top it up does no good either - it empties out of the slave cylinder almost as fast as I can fill it!  That is what a 2 year / 5k mile old IL Motorsport slave cylinder looks like - all rubbers totally perished. I guess they're cheap for a reason. I'm also sure that having the exhaust running so close to them does them no favours, even if I have a little heat shied in place. I ordered a genuine Mazda rebuild kit, but this happened at the same time as the BMW was also off the road waiting for a new turbo core to arrive so to speed things along I bought a spare IL Motorsport one from a friend who no longer has his MX-5 and fitted that. I'll rebuild my one (and see if the mazda rubbers kit works with the IL Motorsport casting) in due course. Refilled it by syringing oil in through the slave bleed port instead of from the top, which seems to have made it a lot easier to bleed. It didn't take too long and my biting point is better/higher than it used to be. Next up standard oil and filter service. That oil filter is an absolute pain to get to, but other than that it went fine.  Then on to knocking another niggle point off the list - when going down a hill, and decelerating in gear through ~3k RPM, some part of the engine would snub against some part of the body and cause a very loud noise and vibration that could be felt through the drivers feet. I've spent ages looking for witness marks trying to identify the contact point, but have never succeeded. I decided to take the easy route, and simply fit a spacer to the right hand engine mount to give the engine a little more room to move.  Gave it a coating of Bilt Hamber S50 as a quick bit of protection, then slotted it in on the engine side of the engine mount, as you can just about make out below:  This lifted the engine around 10mm, it's a bit tight to the bonnet, but doesn't appear to hit it, but more importantly it has fixed the snubbing issue Still can't see any witness marks though... The spacer has also given me more clearance to adjust the alternator belt tension, which is now nice and tight and no longer squealing on startup. The alternator is worryingly flexible on it's bracket though - I have a strong suspicion that's the root cause of the self loosening belt.  Unfortunately it shares a bracket with the engine mount, so pulling it off to improve it is a big job, including supporting the engine and removing the exhaust headers.. One final change to the car - which may seem silly, especially as it doesn't really show in photos - was to replace the weather strips on the windows. The old ones were quite rough and stained and somehow managed to bug me more than all the pink mismatched paint.  Old ones unclipped, gave the hidden surfaces a good clean, then new ones clipped in. Small change but they make the car look far smarter! You'll have to take my word for it though, as I didn't actually get a picture of the new ones. MOT coming up in the next week or so - fingers crossed! |
| |
|
|
|
|
|
May 12, 2022 16:24:16 GMT
|
|
Good luck with the MOT. I was recently in a similar position with my mohair. I gave it a reasonably good scrub clean a year ago which looked great when wet but all the stains came back when it dried. Like yours mine is coming unsticthed in places and the window is cracked so I happily took one of those nylon brushes you stick in a drill to it with Johnson's Baby Bath. Came up really nicely and an awful lot of green come out of it. Sealed it with Fabsil and it's been pretty good since.
|
| |
Project Diary1975 Viva / 1988 T25 Camper / 1989 Mini / 1991 MX5 / 1992 Mini / 1994 Saab 9000 / 1997 Saab 9000 / 2008 Saab 9-5
|
|
Annoyingly, as I bought the car as a non runner I never had it weighed before I started the conversion, but on the plus side, we have a very similarly specced, same year MX5 to compare against that had been weighed here a few years previously.
Then the clutch pedal goes straight to the floor with zero effort. 