MazdaManiac

iTrader: (3)

Possibly. I will try it as soon as I can. It is an easy mod, but It requires that you remove the outlet pipe to avoid having shavings go into the turbo when you drill and tap the fitting.

Both sides are too small. As soon as I can locate a compressor map for the TD06-18G, I'll see what the possibilities are.


With a properly sized turbo, this is mostly true. However, mine will try to hit the full boost target a a touch of the throttle. It is, IMNSHO, undrivable.

rkostolni

For anyone else worried about this and considering upgrading their turbo, here is an email I got from a Bullseye dealer describing their analysis of failed Greddy turbos.

>>
Ryan

I talked to you yesterday. Dave over at Bullseye Power forwarded me your email. I worked at Bullseye, but now I have my own. I am over at The Turbo Trader.

I am guessing that you have been talking to Chris about his Greddy upgrade. From what we saw on his turbo the front side of the thrust bearing was all beat up, which shows that the compressor wheel was surging very badly and they was the main reason that the turbo failed.

The turbo did show other signs that there was an oil problem that could have shortened the life of the turbo. It looked like oil had been backing up in the lower portion of the center section. It did not seem to be backed up enough to blow out the seals, but it was enough that the oil was starting to choke and block off some of the passages in the center section. From what I have seen of the Greddy turbo kit the oil drain has very little to be desired. I can't imagine that by draining the oil back into the oil pan were you drain the oil from to change it can not be the best idea.

These are just ideas. We do not have a car that we can really do some major testing on.

I facts that we know is that the compressor wheel was surging and destroyed the thrust bearing. The other thing is that the oil drain is barely adequate.

The turbo upgrade solves the main issue which is the surge by using a 50 trim compressor wheel that can handle the lower amounts of air flow, but also be able to hold boost up till redline. As far as how much boost it can hold is just a guess right now. We are almost 100% it will hold well over 15psi to red line. This turbo can handle up to 480 hp at 25+psi. It should be able to handle almost any horse power range that most people would want to through at it. If some one needs more power and has the supporting mods we can do other compressor wheels if needed.

Let me know if you have any questions.

Thanks,


<<

MadDog

Just a reference point: my GReddy turbo has about 16,000 miles on it right now. The first 12k or so it was making ~280whp. Since I got it, I haven't run the boost as high as the first owner did.

rkostolni

It depends on how you drive it as to whether you drive the turbo into surge. If you are often at higher rpms with more throttle, it will probably not be driven into surge. I spent alot of my time in boost, but at low throttle positions. Especially when I had the emanage because to much throttle caused it to run shitty.

The less throttle or lower rpms result in less flow, but the boost is always maxed at the wastegate setting. If the turbine was bigger it probably wouldn't be as much of a problem because it wouldn't have enough energy to spin up the turbo unless you were full throttle, or close to it. But the combo of the small turbine and the location of the wastegate line results in the possibility of high pressure at the compressor outlet and very little flow.

two rotors

If all else fails ,look at the way Mazda did it originally!Indeed the port for the wastegate is right on the discharge.But the really damaging surge condition occurs when you snap the throttle shut and air is forced back into the discharge side of the compressor.Your blow-off valve or recycle valve needs to act fast and is perhaps more important in preventing a surge condition than the wastegate.
My recycle line opens on vacuum from the port where Greddy take the wastegate signal.

MazdaManiac

iTrader: (3)

I'm going to have to re-think my position on the Greddy turbo being too small.
I am looking at the compressor map for a TD05-16 right now and it is, mathematically, the perfect turbo for the RX-8. I would imagine (though it is highly possible that this is not so) that the TD06 version would actually flow even more. The TD04-18 is equally good.
Has anyone found a map for the TD06-18G yet?

Here are the maps for these turbos with the performance line for the Renesis drawn in. I based this on 9 PSI and a max airflow of 370 CFM.

ttt

so the greddy turbo is the perfect one for the rx8? is that what u saying?

MazdaManiac

iTrader: (3)

You need to re-read my post.

zoom44

please explain jeff for those of us who arent used to looking at compressor maps

rkostolni

Isn't it more likely to be be the turbine side being too small, or having too small an A/R causing the turbo to spool so easily?

dannobre

iTrader: (3)

Here's the best I could do A TD05H-18G Should be very close

MazdaManiac

iTrader: (3)

Well, the compressor map has two axis:
The vertical axis is the pressure ratio which is the amount of boost expressed as a multiple of atmospheric. 1.0 is atmosphere (~14 PSI or zero boost). A pressure ratio of 1.4, therefore, is 19.6 PSI or 5.6 psi of boost. I use 9 PSI of boost as my reference when sizing a turbo because that is, IMHO, optimal for this motor since it runs out of fuel at that point. That is a pressure ratio of about 1.6.
When you compute your pressure ratio, you also need to take into account the losses through the intercooler. I usually use 1.5 PSI. This would be the difference between what the turbo puts out and what you actually see at the manifold. This means you need a pressure ratio of 1.75 or so to actually see 9 PSI in the manifold.

The horizontal axis is flow. That is the amount of air going into the engine. This is a function of the relationship amongst displacement, RPM, outlet or intercooled temperature (which affects charge density and is a function of compressor efficiency which can be as high as 75% to 85% for a turbo and one of the reasons it is better than a supercharger at 65% or worse, but I digress) and the estimated volumetric efficiency of the engine which ranges between 70% and 100%. Generally, you assume about an 85% average volumetric efficiency (Ve). The N/A flow rate of the Renesis is about 275 CFM at 9200 RPM. At a pressure ratio of 1.6, the Renesis will consume about 400 CFM of air assuming the IC will cool the air down to 130°F or so from its peak of 210°F out of the turbo.
The flow rate is often expressed as lbs of air (a mass number) instead of a flow number like CFM because this is better reflective of the effects of temperature. 400 CFM at 200°F is a lot less air than 400 CFM at 110°F.

Now, the concentric circles on the compressor map show the ranges of efficiency for the turbo at different flow rates and pressure ratios. Using the map that Dannobre provided (which may or may not be good since it is deduced from the existing map of another turbo), you can see that the turbo operates at its highest efficiency (77%) at the pressure ratios and flow rates in the center circle. This turbo is really happy at 14 PSI of boost and 350 CFM of flow. At WOT and 400 CFM, you should be able to put 16 PSI to the Renesis manifold easily! This all falls in the inner efficiency circle. As you move out to the concentric circles, the turbo gets less and less efficient, but still produces boost. This means it is producing more heat to make that boost. You can see, this works against you in several ways.
The turbo doesn't like trying to flow more air at the same boost or more pressure at the same flow. But it also doesn't like making less pressure at the same flow or less flow at the same pressure. If you go off the last line to the right, you are no longer able to force any more air into the manifold without superheating it. If you go off the last line to the left - that is surge. Both are reversion - where air is trying to go back into the turbo. The low end - surge - is the dangerous part because it can actually stop the compressor from spinning quite suddenly which damages the bearings.
This is when the turbo is trying to produce boost with very little flow. Though our little turbo is quite happy making 9 PSI at 400 CFM, it will not be so happy at 9 PSI with only 100 CFM going by. This means your turbo will be surging if you try to make it produce 9 PSI at 2300 RPM.

Now, to use this information, all you do is select a bunch of arbitrary RPM points and express them as a linear percentage of full flow and plot those points on the map. If they all fall all in the better parts of the map, you have the right turbo. If not, you have the wrong one.
Here is an example of a bad one, the T66:



Here is an iffy one, the KKK K26 (IIRC, this is the turbo on the VW and Audi 1.8T)



The problem here is less evident. It will work, but it is not really ever in its most efficient point.
Now, we look at the TD05H-18G.



As you can see, it should comfortably produce the desired boost across the RPM range with no problem.
What I'm not sure of here is what the differences are between the TD05 and the TD06. I suspect it has something to do with the turbine efficiency which is a whole other discussion.

Japan8

... so. Where does this leave us in terms of the issues at hand?

MazdaManiac

iTrader: (3)

Well, the first thing is to control the out-of-control spooling.
I really think that referencing the wastegate to the compressor outlet will do this.
The additional control of a boost controller will further this effect by allowing the signal to go unimpeded when there is vacuum in the manifold.
Having the turbo partially free-wheeling when the manifold is not in boost will relax the throttle response a little and eliminate the surging. It will also allow those with atmospheric-venting BOVs to release the tension on the setting screw a bit since it will not be under so much stress to blow open all the time. This will further relive the surge without adding to the stalling issues.

The other day when I was trying out one of MadDog's new turbo maps, I ran into a problem where, under light but constant throttle, the mix went to an AFR of 10:1 or richer. After thinking about that on the plane ride home, I realized that it was the crazy boost that was probably present in the intake tract opening the BOV and the MAF was reading all kinds of air that wasn't actually going into the motor.

Where this doesn't leave us, however, is explaining why some (or most) people are loosing boost at higher RPMs.

I won't be back to my RX-8 for a few weeks, so I can't investigate any of this for a while. Hopefully, someone will take up the gauntlet.

MadDog

Hmm... Interesting. I have an HKS SSQV which functions on alterations in pressure. I guess that might explain why I don't have the problem you saw.

MD

MazdaManiac

iTrader: (3)

I also have the HKS super-sequential.
It work by a comparison of the vacuum signal and the available diaphragm pressure. If there is more pressure behind the diaphragm than in front of it AND there is a vacuum signal, it will open. All BOVs will open under those conditions.

KJ238

Yes yes.... i have this problem... it happens at -0.2 up to +0.2bar.....

It helps when i lower the amount of fuel at +0.2 bar.. but not a lot.... I can't do anything about - boost though

PS. I use the SQV too

MazdaManiac

iTrader: (3)

OK. Well, try what I suggested - install a vacuum fitting on the outlet pipe of the turbo and plumb your wastegate signal there and see what happens.
Your set boost will appear lower, but you can increase it with your boost controller.

Moostafa29

This is a little off topic, but MM...that is the best avatar I've ever seen.

zoom44

yes unfortunately i had to get rid of it- against forum rules. try another Jeff.

Moostafa29

Damn....

MazdaManiac

iTrader: (3)

Since when is Willie Nelson against the rules?

zoom44

jef thanks for the great compressor map explanation too. really appreciate it.

but the avatar again had to go

MazdaManiac

iTrader: (3)

Now no one has avatars? What was wrong with the last one?

rkostolni

Well I pulled down my turbo Saturday. There was a good amount of oil in the housing and the piping, but other than that the turbo looked to be in good condition. It spins freely with no scraping. There is no in and out play, there is a very small amount of up and down, probably <1/8in. Is there any up and down play permitted or is it toast?

Also I found what looks like a coolant leak from around the intake manifold. Not to happy about that. Can anyone tell me if this will turn into anything more serious than the occassional need to add more coolant? Here's a few pics of it.