Mk1 Cavalier

He's not much of a drinker but I'll buy him a couple of packs of backy when I'm in Turkey next month 😂 Hahaha, loads of room for TRUMPETS. Not a turbo hahaha, stop it 😂

So, brakes. Again. The setup I had after the booster delete was just awful really. Not good, pedal too hard, everything a bit janky, it didn't leak but that was about all it had going for it.


I spoke to Zed. who helped me with a load of questions I had re: converting to dual master cylinders, how to use a balance bar, and other considerations. Thanks mate! It helped an awful lot. So I decided to bite the bullet and just do it. I will pre-face this by saying another all nighter occurred during this a few days ago, and it really was a lot of research, reading, measuring, cutting etc. to get this done.


So we start off with the balance bar:



This is a balance/bias bar from Compbrake. Doesn't cost much, and in all honesty has done the job perfectly well.

Had to get a couple of holesaws, the smaller of which to cut the pedal. Due to several different space constraint reasons I couldn't really raise my pushrod pivot point by much more than 13mm, so I marked 13mm up on the pedal from the original clevis lock pin point, drilled a pilot hole, and got the pedal cut for the balance bar centre tube.



A good fit:



Why this far forward jutting out and not further back in the pedal? I didn't want my pushrods to be too long and when measured and mocked up this was the best spot for it.

Not exactly the most godly welding you've ever seen, but that isn't going anywhere I can assure you haha.



The inside didn't warp and the centre spherical bearing still slid in smoothly.

So, onto the master cylinders. I had approx. 4.25" of space from the firewall to my cylinder 4 trumpet. That really isn't much. One dual circuit master cylinder was totally out of the question. So 2 single port master cylinders were all I could fit. I looked at various options, weighing up quality and cost. The Wilwood compact masters seemed like an absolute no brainer to me, really inexpensive, and only 4.1" long from firewall mount point to the end of the cylinder. Just small enough.

Went with 0.625 bore for the front, and 0.70 bore for the rear, seems to be around what most resources recommend in general for a non-servo dual master brake setup.

Now onto the pedal. It's a mk2.5 MX-5 brake pedal box from the car I sourced the engine/loom and just about everything else from. Well it was once. I already lopped off the throttle pedal portion a while back. Bolt pattern on it to mount through the firewall was 60mmx80mm, too big for what I needed now. With the balance bar set up with the clevises in the recommended position (just a few mm away from the centre tube on both sides) I needed the master cylinder through bolts to be 67mm apart, and the master cylinders use what appears to be a standard bolt spacing of 57.1mm for their 2 mount holes. So I welded up the original holes in the pedal, and drilled out new ones. Again 13mm higher, I wanted the pedal box in the exact same position in the car, and this would keep the raised pivot point on the pedal in line with the pushrods on the Wilwoods.

I then used a larger holesaw in the centre of each pair of mounting holes, and drilled through to allow both new master cylinders to slot through:



Note the spacers between the masters and the pedal, these were on the original pedal box, but I cut them off to make them a little shorter when I first made the original setup last year. We tacked these securely back in place once we were happy with everything.

They both slot through nice and straight, 67mm apart as required by my balance bar.



As you can see, a lot was cut off of the pedal box, so after this I added various bracing on the sides that angle up to the top, reinforcing the firewall face to the pedal mount part of the box. I also double skinned the front of the pedal box as I had removed the brace points on the inside, just to make it a bit stronger. Felt nice and sturdy.

I needed as much engine bay space as possible, so I binned off the 5mm ali I used before and went with a thinner piece of steel. Much less soft than ali of course, but still works well as a bit of a load spreader plate type thing. Initial test fit:



Of course I had to re-drill the firewall with the new bolt spacing, and holesaw the 2 new holes for the master cylinders. Again 13mm higher than before like the other bits, to keep the pedal box exactly where it was previously.

When we were happy with this I got the pedal mounted inside, made up a sleeve and lock/jam nut for the balance bar, as Compbrake supply a remote adjustor cable for this balance bar separately, which I didn't go for, they don't include a lock nut. Was fairly simple to sort, I just had to go to a nut and bolt store to get a 7/16 20 nut, because apparently every chain hardware store literally only sells metric "M" sized stuff and literally no packs of imperial nuts. Oh well, it was 44p for 2 haha. Used a couple of very large washers between the clevises and the centre tube as recommended by Compbrake's fitting documentation, seemed to help a lot with keeping the clevises level and behaving, so I'm glad I did that.

Brake lines were of course a total nightmare and took 2 days to sort out. Everything was done and then re-done again, this time it all seals properly, and not for lack of trying. A load of new unions, T pieces etc. Put a lot of hours into those bloody brake lines, thankfully a couple of mates helped me out a lot. We got it all bled, worked out a reservoir solution for the time being, and here we are:





Just about short enough to clear the trumpets:



Still using the MX-5 brake reservoir for now, it has a divider at the bottom so if one cylinder/line fails, the other still has fluid up to the MIN level, which is good for now. But I am going to buy either a large single reservoir with completely isolated dual chambers, or 2 completely separate reservoirs. It will mean re-bleeding, but re-bleeding can't hurt, probably worth doing it again anyway.

The pedal feels so much better. It's softer on initial braking now, so light braking is a lot nicer even with no booster now. When more braking is required the force is obviously higher than before when it was assisted, but it is a manageable amount of foot force, you don't have to brace your back into the seat and stand on it. Completely independent master cylinders too, so nice and safe, and with enough fluid displacement between them to power any potential brake upgrade I may do in the future. Seems funny running this setup with original front calipers and NOS pads, and rear drums, but in the long run this is the hard part, and it's now done and can be left alone, save for any adjustments to the balance bar, which can be done without disturbing the master cylinder or brake lines at all of course. But god it was a total slog to get it done, and honestly if I never have to deal with a leaking brake line ever again in my life I will honestly be so happy. Horrible job.


Anyway the car is great now, bigger trumpets will be sorted soon ( I measured and have clearance for approx. 110mm trumpets) as well as the heatshielding I mentioned before. No filters are planned for them, this is a £200 engine. If a stone goes in, a stone goes in. We die like men.



I absolutely love this car.

Great to see your perseverance, there's always a solution, just depends on time and money sometimes! Trumpet clearance looks pretty good, certainly a lot of effort to achieve that.

Have you looked into brake upgrades that'll clear the wheels?

Solution = time + money x misery

Was going to say the 13's could prove a challenge but sounds like it's all in hand.

Now you need a spare motor to make a high compression build

Well, the MX-5 had the last few bits removed and has now been sacrificed to the mighty claw:



RIP, was a great car, but it had run its course for me.


Anyway that means this is now officially my only car, the daily, the fun car, the workhorse, everything.



Can't complain really, I did say right on my first ever post about this car that I get bored and need an interesting daily. So I guess I've stuck to my guns with that haha.


Went to a morning classic meet with a few mates. AE86, XR2 rep (1.8 Zetec swap, pretty quick little thing!), MX-5 drift car, and mine in our group.



Anyway, all was good and some nice stuff there as well!


I had started encountering a rich running issue last week. I noticed the car seemed too rich across the board, idle, cruise, and on full throttle. A little odd as the map isn't too bad now on the ECU, and the Lambda Control has some authority to correct any deviations to match the target AFRs. So I got the laptop connected and noticed the AFR the ECU was seeing was higher than what my gauge read (gauge was accurate to actual AFR). Gauge was reading about 13.1 on idle, and the ECU was seeing around 14.5, with a load of noise visible in real time view, it was jumping up and down quite wildly. After some head scratching, chasing and cleaning earths, unplugging things to check for shorts, it turned out it was the main engine earth to the body. I had obviously caused an issue with it when I had grounded my new coolant sensor to the same place. Removed it and re-fitted and everything went back to normal and has been fine ever since. Idle now sits pretty much bang on around stoich as it should:



After fixing that earth, the ECU stopped adding extra fuel and the difference between the gauge and the ECU reading is now 0.1 AFR tops, and no noise, much much better. Temp sensor reads about 4 or 5 degrees high, on laptop the actual number when sat idling when fully warmed up will top out at about 92 degrees. But it's close enough to warn me if it ever goes into the red.


While I had the laptop out I thought I'd have a look at the VVT settings. I couldn't seem to get the VVT solenoid to actually work before, I've essentially been running with no VVT. But it's not like VTEC on a DOHC Civic, so the difference with it not activated isn't as immediately obvious. I swapped the solenoid for another one and still nothing, so it wasn't a burned out solenoid causing it.

Poured through all of the ME221 documentation I could, and finally noticed the issue. In the Input/Output (IO) settings I noticed that the VVT Solenoid wasn't present in any of the fields. Meaning that there was no setting in my map that actually told the VVT solenoid to do anything. A little bit odd as this was one of ME's own official VVT 01-05 basemaps. I had a look through the user documentation and found that the VVT solenoid was supposed to be linked to the "LS9" output. In my map that is where the idle valve was selected. I no longer have an idle valve, but this goes a way to explain why my idle valve never really worked that well when I first fitted the ECU without ITBs, as that was supposed to be on a different output setting. Anyway I swapped the LS9 option over to the VVT, and immediately it started working.

I tested the VVT function by adding VVT timing at idle, and it shot straight up to around my chosen number. A little bit of messing with the PID settings and it would hit my number, not overshoot, and not oscillate too much. It also made the idle sound hilarious. The child in me wants to leave it with the "Ghost Cam" lumpy idle, but it is stupid and uses way more fuel.

Anyway, after looking up VVT tuning, what others use, recommendations for its use etc. it became immediately apparent that the VVT table on the basemap was pretty horrible. VVT advance across the entire map, all the way down to the overrun fields, it didn't really make any sense. This was my first attempt at clearing it up and making it more sane:



You can see here the Current Advance sat at 40, which is what I set it at for idle for test purposes. All working really nicely. Now the main thing will be to remove the 25 degree advance lower down, it really isn't needed at such small throttle inputs, from all the discussion I could find on it no one suggested mapping the VVT for low throttle cruise. You'll also notice it taper off quite heavily towards the redline, it seems the VVT on these is most effectively used to boost mid range performance.


Obviously it is essentially a stab in the dark doing this not on a dyno, but now it is working I want it to work sanely, so cleaning up the table to at least get it ballpark seemed a good idea. Going to pull out the VVT advance for lower throttle inputs, and mess with the higher regions a bit. Also going to smooth out the regions where it goes from 0-20 etc. so it has smoother activation state, not just jumping up immediately to higher advance. From the driving I have done it feels fine, I can't say the midrange feels crazy different, but it certainly hasn't hurt at all. It's been good fun to play around with, and I'm glad that I've got it working.

Yeah for sure, I'll probably switch it back off and see the difference. I've cleaned up the VVT table a bit more now (less advance at low throttle) and it's definitely a bit nicer.

That looks great for fore and aft shifting but flimsy for lateral shifts. Could do with two outer rods rather than the one inner.
Just my thoughts…