Jump to content
SAU Community

Recommended Posts

Just now, Birds said:

Power is directly related to torque, son. These engines aren't different enough that if producing the same power there will be a large difference in peak torque. You'd be talking ~50nm at best. The 25 will bring on the torque and power earlier and that's about it. Hardly cog-stripping material. Unless he's button clutching it with a locker and slicks he'll be right.



Tell that to a 7L V8 in an old ute that only makes 200kw but can tow a GTR on a trailer behind it without breaking a sweat. 

RB20s are junk, and their gearboxes are even bigger junk. There is a large difference in low down torque, stop looking at dyno sheets for proof and go test drive the engines back to back. How many people have you seen say that they need to ring a 20's neck above 5k to get it moving. Considering you have a high flow and brag about its response.. I don't think you could live with a 20 even though its "only ~50nm at best" 

If you're gonna convert it to 25, which you won't because you've already made up your mind, buy the box... otherwise whatever.

Completely different engine. Let's keep the comparisons sensible shall we...given we're talking about two engines that, all things equal, are 500cc different. That is why I'm happy to generalise that, with similar power levels, one won't be smashing cogs over the other.

FYI...I have driven RB20s and RB25s, back to back. I agree the former is a piece of shit that takes forever to make power, but this fact is largely irrelevant, because when it does make that power it is nearly the same peak torque being transferred through the gearbox. Load and peak torque are what smashes gears, not where in the rev range that peak is...which is ultimately the difference between a 20 and a 25.

And if we're going down the path of experience...well...I've fitted countless 20DET gearboxes to modified vehicles, from VLs and 31s to S15s and 25DE+Ts. Done a lot of 25DET gearboxes too. It is the far superior gearbox, but as long as you're not smashing synchros and dumping button clutches with slicks, you're unlikely to break the 20DET box with less than 250rwkw in a 25.

Anyway, I can't see Pat putting in a 25, as he doesn't want to overcapitalise on a 5k(?) car. Hence the garden variety detailing jobs.

  • Like 1

Quick question for the IT dudes in here. Revolution IT did the load testing on the Census website and ran it at 150% of the expected users to login at the same time. They charged the government $500,000 to run these tests. Does that seem exorbitant? I can imagine they'd have a powerhouse of a computer and server system but god damn that's a lot of money for something that still failed anyway...

well they failed so yes exorbitant.

 

to load test something that they expect 10 million hits in a couple hours...well you have to replicate millions of users, honestly to do it properly I would expect it to be more.

 

 

 

Completely different engine. Let's keep the comparisons sensible shall we...given we're talking about two engines that, all things equal, are 500cc different. That is why I'm happy to generalise that, with similar power levels, one won't be smashing cogs over the other.

FYI...I have driven RB20s and RB25s, back to back. I agree the former is a piece of shit that takes forever to make power, but this fact is largely irrelevant, because when it does make that power it is nearly the same peak torque being transferred through the gearbox. Load and peak torque are what smashes gears, not where in the rev range that peak is...which is ultimately the difference between a 20 and a 25.

And if we're going down the path of experience...well...I've fitted countless 20DET gearboxes to modified vehicles, from VLs and 31s to S15s and 25DE+Ts. Done a lot of 25DET gearboxes too. It is the far superior gearbox, but as long as you're not smashing synchros and dumping button clutches with slicks, you're unlikely to break the 20DET box with less than 250rwkw in a 25.

Anyway, I can't see Pat putting in a 25, as he doesn't want to overcapitalise on a 5k(?) car. Hence the garden variety detailing jobs.



I agree with everything in this post

I was tempted to see what would happen if deliberately inducing lag by ramping boost late (like a centrifugal charger) would be a way to get 400+kw from a R33GTST box.

Would be interesting to see what the torque curve would be like if you run say 14psi at 4k then ramp it slowly to 25psi at 7k. May be a way of keeping a box going. Should result in a big flat line for torque as opposed to a massive spike then it falling off.

12 hours ago, emts said:

well they failed so yes exorbitant.

 

to load test something that they expect 10 million hits in a couple hours...well you have to replicate millions of users, honestly to do it properly I would expect it to be more.

 

 

 

Pretty much this.. I'm not in the webhosting business so can't explain HOW they'd actually test millions of users at the same time accessing their server(s) I'd guess they'd have a bunch then just use math to multiply the load up to however many users.. 

Lets not forget the failure could be due to hackers etc attacking 'for the lulz'

And also that there's a lot more underlying than just server load.. you've got to have enough network infrastructure and bandwidth to flow all that traffic which I reckon is where it would have failed. Like the Monash. 

Don;t call these guys Hackers, thats giving them way too much credit, it is script kiddies doing a Ddos via a zombie network. (if it was an attack)

 

It's that easy even Birds could do it.

http://www.symantec.com/connect/blogs/renting-zombie-farm-botnets-and-hacker-economy

 

the funny thing is load testing is a very similar process, get a whole bunch of boxes and start loading/posting data to the site. monitor.

 

What is interesting is that this might not actually be a DDos attack just a crappy scoping exercise.

so apparently they estimates the traffic at max 1 million users per hour, so the load testing at 150% would be 1.5 million users.
now lets say there around 10 million users submitting online.
I would guess around 1/2 would have sat down in the evening and gone ok Fk it lets do it.
so 5 million in 2ish hours or much more than they scoped.

they could have Ddosed themselves with legit traffic.

 

that also said doing it in the "cloud" is meant to mitigate this as they can just virtually throw more CPU's at it till traffic is gone. 

 

 

 

3 hours ago, Kinkstaah said:

I was tempted to see what would happen if deliberately inducing lag by ramping boost late (like a centrifugal charger) would be a way to get 400+kw from a R33GTST box.

Would be interesting to see what the torque curve would be like if you run say 14psi at 4k then ramp it slowly to 25psi at 7k. May be a way of keeping a box going. Should result in a big flat line for torque as opposed to a massive spike then it falling off.

Keeping all the bad parts of a turbo and getting rid of the good is a Mohsen spec idea.

Might as well fit a smaller turbo that will deliver the same boost and power much earlier on. The reason you'd be getting 400rwkw from the box is that the engine is seldom going anywhere near that and you've effectively dulled your turbo to perform like a small one, but with more lag. If the box can't handle 400rwkw, the day you take it to that peak is likely the day it will break. Unless the box weakness is its input shaft, 4th gear being a 1:1 is less likely to strip cogs.

Think of it this way - a chair is rated to seat 100kg. Someone 140kg sits down on it, quick or fast, that chair is eventually going to break under their weight. You can break it with 80kg if you jump on it (ala dumping clutch).

A turbo ramping up fast isn't quick enough to snap teeth; only the torque limit of the box or a sudden surge from standstill will do that.

2 hours ago, UNR33L said:

Pretty much this.. I'm not in the webhosting business so can't explain HOW they'd actually test millions of users at the same time accessing their server(s) I'd guess they'd have a bunch then just use math to multiply the load up to however many users.. 

Lets not forget the failure could be due to hackers etc attacking 'for the lulz'

And also that there's a lot more underlying than just server load.. you've got to have enough network infrastructure and bandwidth to flow all that traffic which I reckon is where it would have failed. Like the Monash. 

Oh I forgot to...carry the 1

tumblr_lkhcjft2j91qztjn5o1_500.png

  • Like 1
16 minutes ago, Birds said:

Keeping all the bad parts of a turbo and getting rid of the good is a Mohsen spec idea.

Might as well fit a smaller turbo that will deliver the same boost and power much earlier on. The reason you'd be getting 400rwkw from the box is that the engine is seldom going anywhere near that and you've effectively dulled your turbo to perform like a small one, but with more lag. If the box can't handle 400rwkw, the day you take it to that peak is likely the day it will break. Unless the box weakness is its input shaft, 4th gear being a 1:1 is less likely to strip cogs.

Think of it this way - a chair is rated to seat 100kg. Someone 140kg sits down on it, quick or fast, that chair is eventually going to break under their weight. You can break it with 80kg if you jump on it (ala dumping clutch).

A turbo ramping up fast isn't quick enough to snap teeth; only the torque limit of the box or a sudden surge from standstill will do that.

That's the thing though, most dynos have *peak* torque at ... I dunno, 4k? Usually when the car hits full boost, middle of the range.

If you backed power off in that range, your torque would not look like that. Instead of being 620nm at 4k then bleeding off to 500, you could have it peak at 500 and stay at 500 the whole way through the rev range.

Look at most people's dyno sheets, the max power figure is not where the most torque is.

At least in my car having huge power at 4k only rips tyres anyway. I'd probably result in a car which is faster through being easy to drive, and won't smash gearboxes. I am aware that losing traction means driveline stress wouldn't happen anywhere near as bad, but usually 33 boxes die from stripping gears. Under load. Usually 3rd. Usually when cars are on full boost/torque and not losing traction anymore.

 

So my idea was simply to spread that out over the rev range instead of having it all come in hardcore when the turbo spools up.
Going a smaller turbo isn't an option for me, my car has the tendency to choke the living shit out of anything smaller than a GTX3582R. I had to run a GTX3076r with a limit of 5700rpm to avoid the car melting itself.

 

Ideally you'd just have a gearbox that can handle the torque. But rarely do we get exactly what we want in modding cars :P

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Similar Content

  • Latest Posts

    • Yea that’s why I said ima test them with multimeter and see the reads.
    • Only at idle. Isn’t a problem when rev it seems.
    • @Haggerty This seems silly to ask, but are you confident in your ability to tune the Haltech?  
    • Next on the to-do list was an oil and filter change. Nothing exciting to add here except the oil filter is in a really stupid place (facing the engine mount/subframe/steering rack). GReddy do a relocation kit which puts it towards the gearbox, I would have preferred towards the front but there's obviously a lot more stuff there. Something I'll have to look at for the next service perhaps. First time using Valvoline oil, although I can't see it being any different to most other brands Nice... The oil filter location... At least the subframe wont rust any time soon I picked up a genuine fuel filter, this is part of the fuel pump assembly inside the fuel tank. Access can be found underneath the rear seat, you'll see this triangular cover Remove the 3x plastic 10mm nuts and lift the cover up, pushing the rubber grommet through The yellow fuel line clips push out in opposite directions, remove these completely. The two moulded fuel lines can now pull upwards to disconnect, along with the wire electrical plug. There's 8x 8mm bolts that secure the black retaining ring. The fuel pump assembly is now ready to lift out. Be mindful of the fuel hose on the side, the hose clamp on mine was catching the hose preventing it from lifting up The fuel pump/filter has an upper and lower section held on by 4 pressure clips. These did take a little bit of force, it sounded like the plastic tabs were going to break but they didn't (don't worry!) The lower section helps mount the fuel pump, there's a circular rubber gasket/grommet/seal thing on the bottom where the sock is. Undo the hose clip on the short fuel hose on the side to disconnect it from the 3 way distribution pipe to be able to lift the upper half away. Don't forget to unplug the fuel pump too! There's a few rubber O rings that will need transferring to the new filter housing, I show these in the video at the bottom of this write up. Reassembly is the reverse Here's a photo of the new filter installed, you'll be able to see where the tabs are more clearing against the yellow OEM plastic Once the assembly is re-installed, I turned the engine over a few times to help build up fuel pressure. I did panic when the car stopped turning over but I could hear the fuel pump making a noise. It eventually started and has been fine since. Found my 'lucky' coin underneath the rear seat too The Youtube video can be seen here: https://www.youtube.com/watch?v=uLJ65pmQt44&t=6s
    • It was picked up on the MOT/Inspection that the offside front wheel bearing had excessive play along with the ball joint. It made sense to do both sides so I sourced a pair of spare IS200 hubs to do the swap. Unfortunately I don't have any photos of the strip down but here's a quick run down. On the back of the hub is a large circular dust cover, using a flat head screw driver and a mallet I prised it off. Underneath will reveal a 32mm hub nut (impact gun recommended). With the hub nut removed the ABS ring can be removed (I ended up using a magnetic pick up tool to help). Next up is to remove the stub axle, this was a little trickier due to limited tools. I tried a 3 leg puller but the gap between the hub and stub axle wasn't enough for the legs to get in and under. Next option was a lump hammer and someone pulling the stub axle at the same time. After a few heavy hits it released. The lower bearing race had seized itself onto the stub axle, which was fine because I was replacing them anyway. With the upper bearing race removed and the grease cleaned off they looked like this The left one looked pristine inside but gave us the most trouble. The right one had some surface rust but came apart in a single hit, figure that out?! I got a local garage to press the new wheel bearings in, reassemble was the opposite and didn't take long at all. Removing the hub itself was simple. Starting with removing the brake caliper, 2x 14mm bolts for the caliper slider and 2x 19mm? for the carrier > hub bolts. I used a cable tie to secure the caliper to the upper arm so it was out of the way, there's a 10mm bolt securing the ABS sensor on. With the brake disc removed from the hub next are the three castle nuts for the upper and lower ball joints and track rod end. Two of these had their own R clip and one split pin. A few hits with the hammer and they're released (I left the castle nuts on by a couple of turns), the track rod ends gave me the most grief and I may have nipped the boots (oops). Fitting is the reversal and is very quick and easy to do. The lower ball joints are held onto the hub by 2x 17mm bolts. The castle nut did increase in socket size to 22mm from memory (this may vary from supplier) The two front tyres weren't in great condition, so I had those replaced with some budget tyres for the time being. I'll be replacing the wheels and tyres in the future, this was to get me on the road without the worry of the police hassling me.
×
×
  • Create New...