View Full Version : Samoht's 1996 Mazda RX-7 Type RS

February 23rd, 2014, 02:34 PM
On Friday I went to Snetterton for a trackday, with the Mazda Rotary Club, (mostly RX-8s).

Initially the track was wet, so it was a matter of feeling up to the limit of grip at both ends in the slower corners. The FD controls its weight so well, and the steering is so natural, that it's really quick and easy to catch slides - the S13 used to tend to lurch about a bit and be a handful to gather up, the FD just maintains its poise and is very controllable. Once I missed my braking point for the first corner and piled in a bit fast. I trail braked to the apex, at which point the inside wheels were on dry tarmac and the outside on wet. The car just slipped into about 10 degrees of oversteer, and came back really easily.

Had a little run with a 996 GT3 which was being rather tentatively driven on the patchy damp surface, so I caught him up easily. More surprisingly I seemed to be slowly closing on him on the straights, which I wouldn't expect - perhaps he was in too high of a gear.

I got a brief instruction session, which was really helpful. Apparently I was braking hard, then coming off the brakes too suddenly, causing the nose to 'bounce' up and go light just at the point of turn in. This was causing understeer, so I would grab a bit of extra lock, then leading to oversteer as the front bit again. Anyway he showed me how to brake smoothly and the better lines to take for some of the less obvious corners, and it was amazing how much smoother, faster and more controlled I felt on my next run.

Snetterton was surprisingly a really fun track, despite being a flat airfield - there are lots of early and late apex corners, and a couple of great combinations, including coming off the straight at 130mph into a fast left that then leads straight into a tight 90 right.

The FD drives so, so nicely on track. The engine is smooth and pulls hard from low down all the way round to the redline buzzer, and however hard you push it, you always seem to have options to rotate the car a little more and tighten the line, it never falls into plough-on understeer, it's just always responsive. I think the fairly short wheelbase means more front-rear weight transfer, so the cornering balance is very sensitive to throttle and brake - so it's really responsive, but also perhaps more sensitive to ham-fisted braking. It's hard to describe how sweet it feels on the limit.

The FD has cost me rather a lot in the last twelve months, FMIC and radiator upgrades being part of that, but it's totally worthwhile to be able to drive such a great car on track. Now I need some better brake pads, and to plan my next circuit outing!

February 23rd, 2014, 04:34 PM
You are bringing a tear to my eye. I've still never driven what remains one of my favorite cars of all time save in video games. What you are writing is everything I have imagined!

Have you driven an RX8?

February 23rd, 2014, 06:29 PM
Cool! T1 at Snett has a bit of a pucker factor if you go in to deep or have a little slide, doesn't it? :) I liked the turn onto the back straight, it was a big challenge to carry speed in without sacrificing the exit. Also, bombhole is fun. Lots of grip mid corner as the track banks and shoots up. Palmer is the best of the new corners IMHO.

In short - I'm jelly. :) Would love to drive there again.

February 23rd, 2014, 11:45 PM
Still one of the best Japanese (or any) cars ever made IMO. A true classic. :up:

February 24th, 2014, 05:55 AM

February 24th, 2014, 06:55 AM
I loved the one that I was so lucky to drive.

330whp was a whoosh of fun. :)

November 1st, 2014, 10:20 AM
So, updates...

The above-mentioned Snetterton outing required a brake rebuild, anyway I now have yellowstuff pads and DOT 5.1 fluid.

Then the turbos weren't boosting properly, the solenoids were dead so I had six replacements fitted. Plus the steering was weirdly stiff and squeaky, mainly in hot weather; this had been a problem since getting the car, but finally was traced to a bearing in the shaft, so the steering column has been replaced.

Finally Jason (at Super 7, the rotary specialist I use) said that although the car was making just over 300hp on the dyno, it was at 95% injector duty. I was surprised at this, since it had been mapped by someone else who is very highly regarded, but perhaps something had changed to push it up since then. Anyway I saw 91% on the Apexi in a quick third-gear blast, so agreed to a new fuel pump and FPR to run higher fuel pressure and bring the duty down. Originally Jason suggested larger secondary injectors, but that was going to be super-expensive with a new fuel rail too, and the fuel pressure increase was enough to bring duty down into the mid-70s with 300hp.

Finally, after all that outlay it was time for some more trackdays :-)

(a combination of pro photos from Bedford, and my own from both Cadwell and Bedford)

As I mentioned on the GT86 thread, I went to Bedford Autodrome with a Toyota-owning friend. Again the day started wet, and I started out taking it very slowly, finding where the track went. As I built up speed on the wet track, I managed to outbrake myself into a left-hand hairpin; my right foot was still on the brake, but my hands already decided it was time to turn in, in order to make the apex. Result was quite significant rotation, enough to require opposite lock, which surprised me slightly. While the FD rewards good driving, it still gives you a second chance if you mess up, which is why it suits me well ;-)


I had a really good play with an MX-5, through the twisty infield we were both traction-limited in the wet, slithering out of corners, until I passed on the straight.

Building up speed, I came up behind an E46 M3, who seemed keen to play, but I was able to stay behind him fairly easily. Squeezing on the power out of a long double-apex right in third, the turbo torque slung me up alongside, and although subsequent acceleration was pretty much level, my initial advantage told by the end of the straight.


Of course, I then wanted to stay ahead, and totally missed my braking point into a 45-degree fast left. Braking and turning together resulted in sliding wide onto the concrete run-off, but with the car still pointing in the right direction, I just nailed it
and carried on without losing much time.


Luckily the day wasn't too busy, and the wide track made it fairly easy for cars to pass each other. However, an Audi TT RS was spectacularly slow through the chicane, but once out on the straight used full power. I was level under power, but not able to gain on him. Since passing is only allowed on the straights, drivers who dawdle around the corners and floor it on the straights can become a bit annoying. I however learned from what my friend had shown me, carried more speed through the 90mph chicane on the back straight, and hauled alongside the TT at the exit, which persuaded him that it was time to let me through.


All in all it was good fun, and learning from my passenger ride with my friend helped me find where I could carry more speed.


By a coincidence of timing, the Mazda Rotary Club were going to Cadwell Park less than a week later, which I couldn't miss. While most British tracks developed from WW2 airfields, Cadwell is a bit like Brands Hatch in that it's set into a steep-sided, wooded valley, and initially developed as a motorbike circuit. However, it's narrower and even more exciting than Brands, plus you get to use the full loop, whereas at Brands most events only run on the short Indy circuit.


Most people's first reaction on arrival is 'that's a long pitlane', not believing their eyes that the actual racetrack can be that narrow. It's a bit like all your favourite B-road bits stuck together, with just a narrow strip of wet grass between the tarmac and the barriers. It's the opposite of a Tilkedrome, in essence.


The challenge is that the track just never stops changing its curvature, writhing like a snake over the bumpy landscape; usually the bend tightens just over a blind crest, or perhaps flicks back in the other direction, just for fun. Play it safe and you will stay on the track, but miss out on both speed and satisfaction. Once you start to get it right, you link up three or four apexes in a single, flowing, fast line, the tarmac unrolling in just the right place as you crest each brow. As you can probably tell, it's my favourite track. It also has a yump, and a section of switchbacks that crest a couple of brows.


Fortunately I got some instruction mid-morning, my instructor turning out to be Adam Morgan, a successful BTCC competitor who does instruction during the week. This was super-helpful in learning the lines, and as long as I was able to concentrate on doing as I was told, I lapped faster and faster. In the afternoon I had a couple of good runs giving passenger rides, putting the instruction into practice. As the other guys in the group were driving RX-8s, some of them were pretty impressed / frightened by the acceleration of the FD, 120mph on such a narrow track gives a pretty good sensation of speed.


The RX-7 is progressive and predictable, just shading into slight understeer in high speed corners with the power on, the steering indicating when the front tyres are at the limit of grip. At lower speeds the back can ease out under power, but perhaps I was unconsciously leaving a little more margin due to the lack of run-off, as I didn't get seriously out of shape all day.

Some other participants had more of a no-holds-barred approach to the track, resulting in a Seven-alike, two MX-5s and an RX-8 hitting the barriers during the day - getting on for a ten percent attrition rate! Still, I think at least some of those were deliberately going all-out, knowing that their cars are relatively cheap to repair or re-shell.

I happened to be spectating when one MX-5 came unstuck; before he came into view I could hear his squealing tyres, so was ready for him with my camera:

to my relief the driver was fine, and happy to have a copy of my photos afterwards ;-)

November 1st, 2014, 10:37 AM
To end where I began, the RX-7's turbos started dumping oil into the air inlet at Bedford, so I'm now looking at turbo rebuild options. Jason was nudging me towards going for a not-too-big single turbo, however I'm not keen on losing the response of the sequential twins, so will hope to get them rebuilt.

November 1st, 2014, 10:43 AM
What about the 99+ (Series VIII) twins? They're supposed to be more responsive yet provide better top-end, and (at least in the US) they're cheaper.

November 1st, 2014, 01:10 PM
I don't know what I am taking about, but it seems like all the development in turbos should give you lots of single unit options - between ball bearing cartridges and twin scrolls and variable vanes you can get the responsiveness without losing top end, and save weight and complexity in the process.

November 1st, 2014, 06:37 PM
20B swap! :D

November 1st, 2014, 08:24 PM

I feel that something like a GT3071R would be a great option on an FD with no or very modest HP goals. That unit delivers very good response even on relatively small (eg, 2.0l) motors while being capable of big HP - up to 500 - and still being suitable for ~300 or so. Again, I know nothing about turbos on 13Bs, but I've driven Saabs with this turbo and it's very good - little to no lag and a big punch. The weight loss of a single turbo is worthwhile, but it sure seems like a single exhaust path through a properly sized single turbo would inherently yield more power. These types of turbos didn't exist in the '90s hence the sequential approach, but I believe if Mazda did it again today, it'd be a single turbo all the way.

November 3rd, 2014, 02:09 PM
Here's a dyno graph for a standard sequential twin FD. The blue line is torque:
see how it's virtually the same at all revs? That, I think, is what makes it easy to measure out the torque at corner exit, and easy to feather the throttle when the tail starts to rotate, because the relation between your right foot and torque at the rear wheels doesn't change as you increase speed away from the apex.

I honestly don't know how a good modern single would compare, except I don't think it would manage to hit peak torque quite so low down, so would be a bit worse in that regard.

November 3rd, 2014, 03:02 PM
You'd be surprised what a modern turbo can do. Modern turbo technology is light years beyond what they in the '90s - which is why they are suddenly so in vogue. Lag and responsiveness aren't really huge considerations for folks with modern power goals. My Solstice made 1bar at 1800rpm out of 2.0 - less flow and less RPM than a 13b for sure. That dyno shows peak boost at about 2800rpm - I don't know what peak boost is, but probably around 1bar. I don't think that is even somewhat ambitious for your engine.

November 3rd, 2014, 07:43 PM
I know the Turbo'd E36 guys are running 20+ PSI and making strong NA like boost with a GT35R...and 600whp.

November 3rd, 2014, 08:25 PM
The MotoIQ RX7 is a sexy sexy LS3 powered beast. So much want.

November 4th, 2014, 06:29 AM
Peak boost on a factory Series 6 (93-95) FD is 10psi.

V8 swapped RX-7s can die in a fire.

November 4th, 2014, 07:12 AM
No LS7 should ever die in a fire.

November 4th, 2014, 05:42 PM
Peak boost on a factory Series 6 (93-95) FD is 10psi.

V8 swapped RX-7s can die in a fire.

Seriously??? They did sequential turbos for 10psi? Nutso. With that info, I wouldn't even consider rebuilding that setup - there is definitely performance up for grabs here.

November 4th, 2014, 05:49 PM
FWIW it can't be compared 1:1 with piston engines. You can't run as close to stoich on rotaries.

I think the later non-US cars ran more like 12psi.

November 4th, 2014, 08:39 PM
Sure, but flow is flow, and for all intensive porpoises 300hp of air is going to be about the same whether rotary or reciprocating. And I get that more boost on a factory rotary motor could have been a risk not worth taking, and I get that 300hp is a very reasonable (excellent!) number for that car in that era. It's certainly believable there was more on the table and Mazda had no interest in it. Whatever the case. Turbo technology really didn't start to take off til the mid to late '90s, so it's within the realm of belief that by the time Mazda could have built a single-turbo 13b that delivered the performance they wanted the FD was already EOL. Certainly, any early '90s turbo car would have been built with late '80s turbo technology, and full boost at 3000rpm was more or less the target. Delivering it at 2800rpm was sort of an accomplishment. But Saab (truly) pushed development to deliver peak torque earlier - 2200rpm c1993 - and these days turbo cars are expected to deliver peak torque at 1800rpm. In any case, if the goal is 10 or 12psi by 2800rpm with a 300hp peak, the world is your oyster. The GT3071R will deliver something like 25psi by 4000rpm on a 2.0l. The ball bearing Super60 in my old SPG did 20psi by 3200rpm (on 2.0l). I'm expecting the hybrid T3/T4 in the XR4Ti to be in that ballpark as well, having an extra .3l to help. And those are old, old turbos - ~10 year old tech. An appropriately sized twin scroll on a 13b will not have a problem, and will probably shave a few hundred rpm off the torque plateau.