Subaru WRX Forum banner

Turbo Upgrade Advice

28K views 21 replies 5 participants last post by  c2bcoolwrx 
#1 ·
I have been researching different turbos lately, and have decided I want something quick spooling that can hold power into higher rpms. I was looking at the TD04L 19T upgrade from Blouch. I really like the results people are getting, especially for the price; However I feel that it might not be the best option for my 2.5L. What do you guys think? Would I be better off with something like a VF39 or VF35?

Blouch TD04L 19T
Blouch Turbo Inc. - Our performance turbo upgrades and rebuilds

TD04 19T Upgrade Discussion
TD04 19T Upgrade Discussion - NASIOC
 
#3 ·
Yes, the VF34 is ball bearing and quicker spooling than it's VF30 non-ball bearing equivalent. However, all VF series turbos are cheap, inefficient, and slow spooling compared to mitsubishi turbos (like the TD04) which are in turn equally cheap, inefficient, and slow spooling compared to garrett turbos.
 
#9 ·
I don't yet, but I plan on doing it. That's good to know.

Personally, I like the variety of mitsu turbos for our car. They're economical, relatively efficient, and come in a variety of sizes with the subaru layout.

VF turbos are not rebuildable, inefficent, and have a limited selection. They're also not any less money than mitsu turbos.

Garrett turbos are very expensive and don't work with the subaru layout

(and when I say layout I'm talking 02-07 WRX + 04-10 STi)
Alright. Well said. I think the TD04 19T is the best choice for me. It's cheap and easier to do. And compared to a VF it seems to be better in the sense that it's rebuildable and more efficient. A Garrett isn't what I want anyways, and its out of my price range.
 
#8 ·
Personally, I like the variety of mitsu turbos for our car. They're economical, relatively efficient, and come in a variety of sizes with the subaru layout.

VF turbos are not rebuildable, inefficent, and have a limited selection. They're also not any less money than mitsu turbos.

Garrett turbos are very expensive and don't work with the subaru layout

(and when I say layout I'm talking 02-07 WRX + 04-10 STi)
 
#10 ·
It's an underused option, particularly for the 02-05 which benefit from the smaller TD04 housing. I think the EJ255 in your car Mike benefits from a bigger turbo based on the TD05 housing, you might want to price out a 16G or 18G. Note though that you'll need fuel and cooling upgrades to deal with any of these options.

I mean, even cramming in the 19T wheel into the TD04, you're still going to push more air through it than it can stay ahead of at 3000RPMS let alone all the way to redline. You've definitely smaller than the balance point between lag reduction and peak power. Course, peak power is overrated when you're not tracking the car.

I'm sorry to confuse you. There are a lot of "right" answers to the question you asked.
 
#11 ·
I was wondering about that. I had read that a 2.0 would benefit better from that turbo. I was originally thinking about getting a 20G, but then I decided I didn't want that much turbo lag. I thought about a TD05 20G as opposed to a TD06, but that's still a lot of lag. I don't see a lot of people with 18G so that makes me think they aren't a great option. And I'm afraid a 16G isn't going to have enough top end power to make me happy.

I realize that I'm being extremely picky. I just want a turbo that spools quick and can hold power to redline.
 
#13 · (Edited)
Well both of those are relative phrases. "Spools quickly" relative to what? Nothing you will look at will spool faster than a TD04-13T. Sucker reaches boost at like 2k RPM on a good tune. Same goes for top end. If you size the turbo to be so big that it makes max boost at, say, 6500RPM, it's going to not make any boost till 5000 RPM or higher.

Personally what I think about is area under the curve. The earlier the turbo generates torque the better but not if torque falls off at higher RPMs. This leads you to a turbo that is what some would call "undersized" but which is large enough to maintain efficiency all the way to redline. You'll actually be below the maximum boost level throughout. After all, if you hit that in the first place, it's inherent that at any lesser RPM, you're still in some degree of lag.

For explanation purposes, here's a dyno graph of my EJ257 with a TD05-16G (regular, "small", etc). Note torque is the top black line and boost is the green line.


Note the slope of the torque curve. The car spools very quickly (it's already at 4PSI where the graph starts at 2.5k RPM) but tops out pretty early. Peak Torque only lasts till 4500RPM. The ideal shape, in my very biased opinion, is one where the peak torque lasts till just before redline. With the shape in mind, you can be a little bit more subjective about what size turbo you think is ideal. Note though that we haven't mentioned target boost or hp/tq levels. If your concern is sizing, these are independent. I would also point to the RPM difference between gears. This gives a rough outline as to the min and max RPM points you should concern yourself with. For instance the gap between redline third and fourth gear (for a constant whatever speed that is) is probably a good measuring stick. You care about torque over that range plus a bit probably. Point is you don't care about power at 600 RPMs.

Note that the EJ257 will spool a little quicker than the EJ255 due to an improved top end. Also the redline is 250-500RPMs further out. This effectively means you'll want a slightly smaller turbo than the EJ257 would.

It's hard to speculate exactly what turbo will give you a given curve just like it is to speculate about what the peak numbers will be but I think the TD05 is probably the right housing. A large 16G is an appealing solution but will probably have some torque falloff, especially with a good 93 octane tune. The 20G will probably hold max torque pretty close to redline but will take longer to spool. The 18G is probably "just right" IMHO.

I'm not the originator of this table but I find it helpful:
TD05 16G
TD05 16G E3 - EJ205
TD05 18G - EJ255
TD05 20G - EJ257
TD06 20G

Go up a level if you want more power at the expense of spool and down a level if you want more response. The TD04-19T is probably a more efficient equivalent of a TD05-16G
 
#16 ·
Looks like the 16g is pretty close to the VF52... Here's a log of my 09 after my tune.

Also makes about 4 psi by 2500, hits up to 17.5 by about 3300, 18.5 by 3500.
I would say the VF52 is a decent bit larger than the TD05-16G, it's just not as efficient. It's very similar to a VF39 which gets compared endlessly to the "regular" TD05-16G. Usually the VF39 makes more peak boost but less area under the curve.
 
#15 ·
Thanks a lot Mosc and Chris. The graphs were wicked helpful. Its good knowing what the most "efficient" turbo would be for my car. I'll just need to save up a little longer for the TD05 18G. It's funny cause I've never even considered an 18G before. Maybe I'll try and find some people with a 2.5L WRX + TD05 18G and see how it's working for them.
 
#20 ·
I guess I should chime in here, being that I have a Blouch TD05H-18G 8cm on my EJ257 hybrid build. IMHO it is an AWESOME turbo on a 2.5L for DDing/AutoX and I love it. Also, 8cm is a MUST on the 2.5L as it really takes advantage of it. For me lag is almost nonexistent. I made 320whp/294wtrq at only 16PSI. See my profile for my full mod list and a link to my dyno chart.

18G is the way to go for minimal lag and w/still a good top end on a 2.5L. Just get it from Blouch, b/c many of the other companies cheap out and use a the smaller TD05 compressor cover which is really meant for a 16G. Blouch uses a TD06 compressor cover which is much better suited for the 18G, it makes more power b/c of it, and is not chocked in the higher RPM ranges.
 
#18 ·
well I don't think the VF52 gets to 320whp, at least not on a dyno I've used. I'm not too far off of maxing the 16G. Like I said, the 16G is actually smaller than the VF39/43/52 but it does have a larger area under the curve.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top