When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I'm looking to put a rear half cage in at some point. Has anyone ever fabricated additional bars which would bolt to the rear cage and then come up to the windscreen like the roof bars in a full cage. These could have the factory hood clamps fitted to then stiffen up the windscreen as if you had a hardtop on. Might be quite nice for a dual road/track car where you can then track it with the roof down but have slightly addtional stiffness and less skuttle shake. Would need to come off when it rains if you don't want to be wet to then put the hood back up. Probably a quick pin arrangement that might work at the connection to the cage. The factory hood clamps would then tension it so it is quick release.
Spoke to my fabricator friend - He has refused to make my initial idea for me. Instead he quickly drafted out a T top arrangement which would be much stronger but still work off of the Mazda windscreen latches so that it is "quick release". Basically it is like and H frame. One side of the H runs across the top of the windscreen and has the Mazda latches. The other side of the H runs along the top of the rear roll bar and is bolted onto the roll bar. 2 bars are then used to make the T top and are bent to tie into the 2 sides of the H. I've got a rough sketch:
The goal is to vastly reduce the windscreen skuttle shake when the roof is off and potentially add some additional roll over protection. Its a quick trackday fixing for a mainly road car that retains the soft top for use on other days. I think it will stop the shake but have no idea on roll protection as this would be subject to the strength of the mazda A pillars etc. which I have no idea about.
Have you though about seam welding the inner doors panels all the way up to the top of the windscreen? This will stiffen all of that up a good bit and would be a lot less work than the bar you are thinking about making. I haven’t driven my car yet since doing this (fingers crossed soon!) but you can tell all of it is much stiffer already just jacking up the car or opening the door. I notice the change in opening & closing the doors the most, but I’m certain the A pillars / windscreen are much stiffer too. I’m a crappy welder / just started to learn here, you can see what I’m talking about. Also, all of this is hidden behind the rubber seal, although I painted mine to keep rust at bay.
There is a YouTube video where someone does this in his Miata and he could shake the windscreen by hand before doing the welding if memory serves. Afterwards no shake.
good luck!
Bottom of door sill, where you get in..
This is up the A pillar on the windscreen...I wasn’t very good at welding on a vertical part, that’s why these are more spot welds than stitch welds...but even this tightened it up.
Was out testing tonight. Running 7psi flat to 5.5k rpm then rising to 8psi at 6k, 9psi at 6.5k and 10psi at 7k. These figures are based on my NA power graph with the intent of holding c190lb/ft right to top end past where the NA power drops off around c.6.5k. It pulls really well.
Did the standard test of 5th gear (6 speed box) 100%TPS to find the boost threshold. Hits the 7psi at 2750 rpm and feels very responsive. Turbo match for car appears surprisingly good initial tests. Hope to get it on a dyno before month end.
I'm looking to put a rear half cage in at some point. Has anyone ever fabricated additional bars which would bolt to the rear cage and then come up to the windscreen like the roof bars in a full cage.
One of the Japanese full cages had a similar arrangement to this - bars down the centre rather than above the doors. I'd check head clearance pretty carefully so you don't brain yourself should the worst happen..
One of the Japanese full cages had a similar arrangement to this - bars down the centre rather than above the doors. I'd check head clearance pretty carefully so you don't brain yourself should the worst happen..
And remember to check it with a helmet on as that brings spaces in a ton.
Quick log from road testing tonight. Flat road - fifth gear (1:1 gear on a six speed), exhaust "hot" from previous pulls. Hits target boost at 2550 RPM.
"Hot" is relative as you can see the exhaust gas temp is low as its only running baby boost as I'm on a sealed VVT engine and pump gas.
Fast spoolin' for sure. Youre going to need rods if you want to turn it up much more, if any. Either that or use EBC to taper the boost up with RPM a bit.
Fast spoolin' for sure. Youre going to need rods if you want to turn it up much more, if any. Either that or use EBC to taper the boost up with RPM a bit.
Yup, I want to make it last the summer. I have a spare engine to build as a project over next winter.
Boost is currently tapered up where torque dropped off on the NA setup: Targetting 6.8psi to 5500rpm, 8psi at 6000, 9psi at 6500 and 10psi at 7000. I then knock off the duty a chunk to soften the curve into the limiter. I've got a shift light on the way as it is impossible to determine the revs by feel as the boost taper feels like it has filled the potential torque drop off at the top end. Current aim is 190 lb/ft flywheel flat to 7000rpm. Over boost cut is mapped against RPM at circa 0.4 psi over target so there is a safety.
There is a lot more than a few weeks in this and all the supporting modifications, learning to map etc. Started the idea early last year.
Virtual Dyno hates Haltech log files. I've messed around with the config file but never got it to work correctly - it thought I had 2500hp last time I tried so maybe something like time is out by a factor of 10. I'm just going to wait until it is on the dyno for actual figures. It was 142 engine, 120 something whp on an eddy current dynojet at sea level with "certified" SAE corrections (whatever that means).
190tq at 7k is plan - I don't know if it is there yet or not. Ignition timing is quite conservative and it could probably be leaned out more at this low boost as shown by the low EGTs - so probably not quite the equation 190tq. Again don't want to push timing until I can see the trends on dyno to see where it is worth it. I do run det cans and knock monitoring on the ECU and believe it is currently happy. At 600miles in it better be!
Yes - Completely flat road and zero brakes for the purpose of putting up a comparable log. The exhaust was "hot" as in the car had been driven spirited ahead of the log so best turbine energy for early spool. A cooler run up to the logging run would have probably yielded just below c3000 RPM spool. No one talks about these subtle differences in response.
I'm currently mapping a correction factor in my boost control based on EGT for exactly this reason.
Crispy cold weather will have also helped air density.
I'll probably be out for a play in it tonight so will try to get you a very clean log and you can try to figure out VD. It recognises the Haltech log, it recognises the RPM. I think it struggles with Time and TPS. I have to go into the columns config file to make it match. I guess I could try this the other way around and go into the log and change the column names to match VD there. Hadn't tried this. VD would be nice to have an indicator of power trends if done on same stretch of road/direction etc. each time.
I can't believe this actually worked! Well done, brother. Can't wait to see what happens next.
Many thanks - I'm still verifying its reliability but it has done 600 miles of hammer so far.
I'm currently looking at the Mitsubishi Secondary Air System holes in the evo exhaust manifold which I put temporary bungs in for simplicity. I'm going to open these out and put in larger pipework than standard and plumb these to a Mitsubishi vacuum operated air bypass valve that is then connected to the intake side ahead of the throttle. I'll control the valve with an additional boost solenoid and the ECU to provide a rally type fresh air antilag system, much as Mitsubishi did in the Group A Evo rally cars.
I already have the SAS valve and a spare boost solenoid valve. First I'm going to wire up the solenoid and plumb up the vacuum signal hose to the valve. I'll then map and test the valve external to the engine (other than the signal line) in the situations I want it to work to validate how it operates before I fabricate the custom pipework and mountings with a friend. Just another science project. This one will be over a much longer time frame as I'm still mapping and playing with the car and main focus is to keep zipping all the little jobs up like air dams, undertrays, filter enclosures, wiring etc. ALS is can also be v hard on an engine and I'm trying to keep it alive through summer so installation testing might be in Autumn.
Quick funny one from mid fabrication a few weeks ago. Me and my friend were quite impatient to see how the turbo performed before we had finished the fab work. Below was our 2 minute wastegate control solution before we headed out for a quick road test. The 2 spring setup yielded c3psi and the 4 spring setup yielded c6psi until back pressure opened it slightly at higher revs.
This was a great motivational exercise and high fives and whoops were exchanged with boost response results. I was always in it for the boost. We then finished off the fab work knowing it was going to be pretty good.
I'll be getting a turbine back pressure sensor into one of the scrolls ahead of the dyno session so we can see how it is performing against boost.