NA tuning - gear ratio and gearbox selection

[mjrstar]

Just a general thread with some musings, as I have been considering an engine upgrade for my pos race car to try and make it go fasterer…

 

But the downside with my would be chosen engine selection has nothing to do with the engine itself or the installation which is all pretty straight forward, and yes the engine would have more HP and torque, and is of a newer era. but is to do with the cost of building what I would deem to be an acceptable gearbox with ratio’s I’d be happy with, and will shift seamlessly at high RPM. (less RPM drop between gears means less “work” for a synchro to have to do)

I realise some people think that a diff ratio change can solve some of these issues, but in my opinion yes this will modify torque output at the wheels per gear, and speed the car will go in each gear, but as far as keeping the engine in the “sweet spot” it won’t optimise the rpm drop per gear.

The other thing to consider, is synchro type, availability of aftermarket support, as newer designs and materials to increase efficiency.

Then clutch selection, if I have an engine that makes 250hp and say 250NM of torque, yes I could put a massive full face clutch disc capable of holding 600NM, but this adds rotational mass that increases workload on the synchro.

There are a few calculators online, the https://www.blocklayer.com/rpm-gear is alright as it gives you a PDF to download (great for ADHD vs just displaying on the screen) but its festooned with ads.

Anyway with some of the NA tuners here getting balls deep into awesome ITB retrofits, VVTI conversions at all sorts of trickery id thought it’d be interesting to see how many others are considering gearing adjustments or at least use a gear ratio calculator to see if their intended gearbox of choice will deliver to their expectations.

[mjrstar]

Example here is my POS with aftermarket 3,4,5 gears

I had to raid the input shaft and 2nd gear from my mates DA Integra with over 300,000km, as the 2nd gear in my existing gear kit was almost the same as the aftermarket 3rd gear.

[Roman]

I've got a program called CarTest2000. It's pretty janky in terms of it's layout etc, but its really great.

Basically you input a car's tyre size, aero, weight, gear ratios, all that. Like so:



then you can run a bunch of performance tests. 

The great thing about it isnt necessarily that it's 100% accurate, so much as that you can isolate just a single variable perfectly, and see how it changes things. 

In saying this, before I'd turned a single spanner on a 1NZ, it predicted the Echo would run a 13.8 and 100mph. Which was not far off the money. 

Then it can output a variety of tests like this





Or like this for comparative testing



 

[Roman]

The other thing that's great, if you want to get a more real life sort of feel about how the car will perform. 
Cars in Assetto Corsa can be adjusted by tweaking the values in text files. That just list things like gear ratios, etc. 
My gearbox has a really really awful shift from 1st to 2nd, and ideally I want to use the 3-4-5-6 gear stack.
However to do this with lots of rpm would requre a 5:1 final drive, or higher. 
Putting the car together in Assetto Corsa, and driving it around Hampton or Taupo (there are really good maps for each of these now!) and it showed that it was just quite horrible with a 5 or 6 to 1 final drive. 
As you'd have to change gear mid corner so much.
4.7 final drive actually seemed to work well, and doesnt hugely compromise the cars ability to trundle along at 100kph etc. 

 

[mjrstar]

I'm all about compromising my 4500rpm+ at 100kph as nowhere I go has a top speed greater than 200kph, and the engine would struggle much past that sort of pace. 

I have got close to rev limit in top gear, and probably would now at Hampton.

Unlike my starlet which was on the limiter in 5th before the kink the at pukekohe, even when I put the taller 3.9 diff. Which was around the 250kph mark which is too much for a kp starlet.

[mjrstar]

That Car test thing looks pretty comprehensive @Roman I wonder how close my car would be.

Civic did a 14.0 at 100mph at os drags, but couldn't sort the wheelspin in 2nd gear it was WAY worse grip than coarse chip seal.

[Rhyscar]

Agree with where you're going here.. Aftermarket gearsets for Hondas are readily available and reasonably priced (I think anyway), for what you achieve in performance. For example, you may need to spend twice the budget on the engine to see the same performance gain. 

If gearkits were readily available at reasonable cost for Toyota fwd boxes, I'd be heading down that route eventually. However, similar to Dave, I've worked out my c60 box gets infinitely better changing it from 4.5 to a 5:1 final drive. For the circa $2k cost, this seems like a cost-effective improvement once my wallet recovers from the final stages of the build.

Alternatively, I think I'd struggle to justify an extra $4-6k on a special gearset, as I think I could find another 10kw with some porting and cams at a lesser cost. Also partially because I haven't found any proven solutions out there. 

 

It's an interesting equation to work out (performance gains * smiles/cost)*proven solution = is it a good idea? (and perhaps a few other variables too)

 

If you don't want to chase power, gearsets can be a cost effective performance improvement, but if extra power is *relatively* easy to find reliably, it may be further down the priority list.

Other option is to drop the same amount of weight with carbon panels (in my experience it also offers some driving dynamic enhancements that can't be measured by numbers). 

[mjrstar]

I guess with the likes of a c60, you have the opportunity to purchase 6 new ratios, and replacement input shaft from the likes of KAAZ, yeah it's costs some coin but then again it's over twice the hardware of a wallet friendly 3,4,5 set for a Honduh.

I reckon it would definitely go hard not even just for what it is but in general terms too. Maybe @Roman could do you a graph to identify the level of hardness?

Harder than 6k of engine mods? Probably not so I get your point @Rhyscar 

whereas you can drop sizable dollars on a honda engine build with what seems like negligible gains from a stock type r B series, for example there is minimal valve pocket clearance to allow for much more lift or duration unless you put new pistons in.

The k series seems like a different animal, with decent gains to be had ( I suspect this is the difference of vvt versus 1980s caveman spec vtec)

 

For the rwd crowd i see there is a j160 6th gear swap that people do depending on what gearbox you start off with.

[kpr]

Some of the problem is peoples approach at  modifying na engines.   your kind or pushing shit up hill to make good numbers without rpm.    but if your smart about it  you can at least retain the power off the bottom and gain a bit of midrange along with top end.  which makes an easier faster car to drive, even with the same ratio's.      good chunk of people still assume that making power na  will loose low/mid for top end gain.  then just run with it when it happens.   which yeah, will send you directly to close ratio box territory,  if you actually want to beat the guy with stock engine. 

some r&d  into the exhaust and intake maybe cheaper than a 6k gearbox.     

[Rhyscar]

When we developed dad's AE82 Corolla 10-15yrs ago, this is the order we did it in. 

1. Lightweight carbon roof, bonnet, boot. Cost approx $8k to build custom items

2. Moderate power level - 140kw, basically stock engine internally, cams etc approx $10k engine

3. M-factory gearkit - this worked great until dad had an elder moment and forgot to put oil in it.. at the time this was around $4k

4. Big power, forged engine, CNC head etc, lots of valve failures, 100hr lifed engine $20k min

5. Quaife sequentially shifted dog box $12k + full rebuild on delivery to fix manufacturing errors

6. Full underfloor aero mods, carbon fibre all panels $fuckinlots

 

It always performed better than expected up until around step 3. Mainly because we started with lightweight stuff first, getting it to 814kg or something silly before adding other mods. I always thought it should have never gone past step 3 as it was an awesomely reliable car that gave almost no problems at that point, but dad was determined to make it the fastest most expensive corolla in history. Currently sits in the shed, not getting used cause it's tricky to get everything running right, and needs a mechanic to get it to run properly. 

That's been my experience anyway. I don't think I'll go past a final drive in the Levin, better off being used like your civic, if not perfect. Generally, cars go faster with new tyres than with the same value in new parts. 

[mjrstar]

4 hours ago, Rhyscar said:

 I always thought it should have never gone past step 3 as it was an awesomely reliable car that gave almost no problems at that point, 

That's been my experience anyway. I don't think I'll go past a final drive in the Levin, better off being used like your civic, if not perfect. Generally, cars go faster with new tyres than with the same value in new parts. 

This is where I'm kind of currently sitting, but gave not really put the car on a proper diet. ( still has electric windows, carpet, power steer etc) 

 

i figue the step from here is exponential expense for something that's currently pieced together from mostly second hand parts and owes me sub 12k all in to basically double or triple that.. I might as well just get a better car than turd polish this thing.

Problems is when this unknown engine decides to expire, finding a replacement 30 year old unit worth installing will be expensive and difficult.

The platform I have lends itself to other options,  k20/24 swap, h2b, b20 etc.. each with thier own pros and cons.

What would be cool is an f20c running the b series box I currently have. Which may be possible by running the engine backwards with a few head scratch areas of course. I like the idea of 240hp 9000 rpm and current gearing.

 

 

4 hours ago, kpr said:

  good chunk of people still assume that making power na  will loose low/mid for top end gain.  then just run with it when it happens.   which yeah, will send you directly to close ratio box territory, 

@kpr I think technology like twin cam,multi valve and ecu tech for sure have made big inroads in this area. Gone are the days of push rod, or SOHC stuff where you really do have to make a compromise. In fact places like Toda, do b series cams specifically to fill the hole in the mid range on the low lobe cam before vtec yo kicks in. And can add close to 1000rpm of usable powerband.

[Hyperblade]

3 hours ago, Rhyscar said:

When we developed dad's AE82 Corolla 10-15yrs ago, this is the order we did it in. 

1. Lightweight carbon roof, bonnet, boot. Cost approx $8k to build custom items

2. Moderate power level - 140kw, basically stock engine internally, cams etc approx $10k engine

3. M-factory gearkit - this worked great until dad had an elder moment and forgot to put oil in it.. at the time this was around $4k

4. Big power, forged engine, CNC head etc, lots of valve failures, 100hr lifed engine $20k min

5. Quaife sequentially shifted dog box $12k + full rebuild on delivery to fix manufacturing errors

6. Full underfloor aero mods, carbon fibre all panels $fuckinlots

 

It always performed better than expected up until around step 3. Mainly because we started with lightweight stuff first, getting it to 814kg or something silly before adding other mods. I always thought it should have never gone past step 3 as it was an awesomely reliable car that gave almost no problems at that point, but dad was determined to make it the fastest most expensive corolla in history. Currently sits in the shed, not getting used cause it's tricky to get everything running right, and needs a mechanic to get it to run properly. 

That's been my experience anyway. I don't think I'll go past a final drive in the Levin, better off being used like your civic, if not perfect. Generally, cars go faster with new tyres than with the same value in new parts. 

In circuit racing, you do some of the above, and then you are now no longer racing with the cars that you used to be around, so just end up lapping the circuit on your own, which isn't that exciting.

And there will always be a car faster then you even when you've spent all that money...

The best moment I've had on track was driving side by side with someone (I trusted to race) pretty much the entire circuit (can't even remember if i beat him...).

One thing that could actually be an investment is driver coaching, the right person can make a very big difference on how you drive the car and therefore speed.

So easy to say all that, however I personally can't stop myself from improving the car, and all the shiny go fast bits are so tempting...

[Roman]

I 100% buy into the idea that finding that sweet spot of price / effort / complexity / failure rate / etc is the main thing that can make a car fun or not.

If you have a car that has dry sump which needs 1/2 hour of oil warming.
A rollcage you need to climb over.
Sparse interior with no comforts.
Ear splittingly loud. and so on. 

You just super narrow down the conditions in which you could go take the car for a fun drive. And then the less you engage with it, the less you feel like it later. and so on.

All just adds up to the point where there's so much admin work needed before you can go for a fun blat up the road. That you just leave it parked up instead.

[kpr]

1 hour ago, mjrstar said:

 

 

@kpr I think technology like twin cam,multi valve and ecu tech for sure have made big inroads in this area. Gone are the days of push rod, or SOHC stuff where you really do have to make a compromise. In fact places like Toda, do b series cams specifically to fill the hole in the mid range on the low lobe cam before vtec yo kicks in. And can add close to 1000rpm of usable powerband.

Its relevant to most / all engines.  the exhaust and intake plays a huge part when comes to n/a tuning, no mater the engine configuration.    unfortunately some of the things that work the best can be hard to manufacturer  or have fitment issues.  so the aftermarket leads people down the wrong path at times.    
Take for example the b series honda.  some big diameter headers and a short intake manifold; will make some numbers up top and get away with your sins at low/mid rpm running on the small cam profile.    But put the correct size header, exhaust and correct length intake on same engine.  It will let you take advantage of the high cam profile at a lower rpm and  make a stack more through the middle. 

 

 

[shrike]

22 minutes ago, Roman said:

I 100% buy into the idea that finding that sweet spot of price / effort / complexity / failure rate / etc is the main thing that can make a car fun or not.

If you have a car that has dry sump which needs 1/2 hour of oil warming.
A rollcage you need to climb over.
Sparse interior with no comforts.
Ear splittingly loud. and so on. 

You just super narrow down the conditions in which you could go take the car for a fun drive. And then the less you engage with it, the less you feel like it later. and so on.

All just adds up to the point where there's so much admin work needed before you can go for a fun blat up the road. That you just leave it parked up instead.

I feel this, but considering  id have to make the car factory to bring it over here, which is pretty challenging as id need almost a crashed car to do it. Also my options to sell up are a partout and I cant do it to myself

And it's gotten really hard to have a modified car in Victoria. Id be lucky to have the car road legal with a half cage.

Need to see where things are at when I come over and rethink based on what's in front of me.

But at least ive got most of the parts to make a call, and part of me is already so far into it I might as well see it through.

Might still be a possibility to get it certed with a logbook and motorsport licence in NZ. Id really loke to get it done enough to bring to Aus and then tinker on it in the shed.

Try and make as many events as I can

I am trying to dial back my build and make it enjoyable. Im really trying to avoid sinking too much into the engine or gearbox currently.

If the K24 ends up being unviable then that will be a rethink.

Can still throw a 7afe and e59f into it lol 😆

[Roman]

19 minutes ago, kpr said:

Its relevant to most / all engines.  the exhaust and intake plays a huge part when comes to n/a tuning, no mater the engine configuration. 

This is an interesting point, because if you get that midrange beefed up, then it might feel like it doesnt need that final drive change, or the close ratio box quite so much. 

Then you get all this mileage out of the car using OEM parts instead of high maintenance dog box or whatever. 

Accumulative wins from getting the basics right, and vice versa.

[Rhyscar]

@kpr you’ve really opened the can of worms with a lot of that development stuff having a dyno handy and the time/skills to make different iterations of stuff and then testing it using actual science. 
 

Most of us will have one or two trips to the dyno, call it good enough and not go any further.
 

But starting to see more and more like you say there plenty of power gains available for fairly minimal input. 

@mjrstar if you could find 10-15% extra power with a bit of effort would a gearing upgrade still be worthwhile? Hmm 🤔 

k/f series route is cool and all, but is expensive these days. No longer wrecker motors. I always really like the b20 setups and reckon they sound the tits with cams etc. perhaps even vvti on a b series?? 
 

Also don’t buy someone else’s half cooked project. Unless it’s already winning races, it’s not worth it (but it’ll be way more expensive!) 

[shrike]

5 minutes ago, Rhyscar said:

@kpr you’ve really opened the can of worms with a lot of that development stuff having a dyno handy and the time/skills to make different iterations of stuff and then testing it using actual science. 
 

Most of us will have one or two trips to the dyno, call it good enough and not go any further.
 

But starting to see more and more like you say there plenty of power gains available for fairly minimal input. 

@mjrstar if you could find 10-15% extra power with a bit of effort would a gearing upgrade still be worthwhile? Hmm 🤔 

k/f series route is cool and all, but is expensive these days. No longer wrecker motors. I always really like the b20 setups and reckon they sound the tits with cams etc. perhaps even vvti on a b series?? 
 

Also don’t buy someone else’s half cooked project. Unless it’s already winning races, it’s not worth it (but it’ll be way more expensive!) 

I can still pull K24s for 300-500 at a Wreckers. But they are alot higher km now

[Roman]

One thing that I think is interesting, is that in our brains we build false equivalencies between various factors.
As in. getting more grip, getting more power, having a more aerodynamic car, etc are all worth the same in how we emotionally or financially invest in them.
Spend $5000 on a carbon bonnet, has to make your car equally as flash as spending $5000 on a new gearbox right?

Yet, in reality. Something like adding 10hp might be fairly trivial, reducing drag coefficient by 0.5cd might cost you $20,000. Or not even be possible. 

You might spend $5000 on a carbon bonnet, that only saves you 15kg. Which has the same net effect as gaining 3hp. haha.

So its interesting to take the same baseline car, and run it through the CarTest program. Even if just for the straight line stuff, results are interesting. 
Below I have setup different versions of otherwise identical car with +10hp, 0.5cd improvement, 50% more grip, and 50kg lighter. As I think these all show realistic possibilities from things like buying different tyres, adding carbon parts, taking out seats, or whatever.
It's interesting to see that adding 10hp wins or equal best in every category except for 60ft time.





I guess what I am saying is, if you adjusted the gains from each of these categories until the results were about even for all of them. 
You'd probably find that one will be impossible (improving .cd by a massive margin) 
Others will be insanely expensive  or detrimental to the overall car (trying to lose 300kg) 
and others are comparatively easy (Spend an afternoon on dyno testing 5x different intake pipes)

 

[Hyperblade]

10 hours ago, shrike said:

I can still pull K24s for 300-500 at a Wreckers. But they are alot higher km now

That's just the start, you want the decent CL7R gearbox so add $2000 and your hoping it's actually in decent condition (most have issues) as they are all high k's these days.

And that's with stock LSD which isn't good enough for serious race cars, so add another $2000 for a plate LSD.

Already up to $4500 (That's just part costs not labour...)

Then you putting in a chassis that didn't originally have it so start adding $$$ from there, it is easy because of all the parts available, but it's not cheap.

Here in CHCH 2.4L na  puts you into the 2L to 3.5L class (which includes 2L turbos and Rotarys including 4wd) so how competitive are you really?

On the flip side you will have a very reliable consistent race car with easy access to parts, and a cheap engine swap if you break it.

Sometimes makes sense, other times not really.