Turbo looks like a great option .... pricing very similar to Garrett but some great features! That size looks near perfect !
Some minor exhaust porting only .nothing on the intakes.
BTW I don't really agree that the 660 compressor is equivalent to a 7670 ..... It's actually closer to an 8374 (in efficiency) in the Pr ranges suitable for a Renesis. With that xona comp wheel .... It has a low trim value which means it will be more suitable for a smaller engine running high boost ...same as the 7670.

Did a bit more wrapping and heat shielding of manifold.
Result : Gain up top from more boost creep.
Bigger (44mm) WG is on the way.

https://cimg7.ibsrv.net/gimg/www.rx8...9f1a00701.jpeg

my reference to it being similar to the 7670 is wrt flow potential which was the context at that part of the discussion

because context matters

in a broader context, it’s better than a 7670 in several key ways, the noted efficiency difference being one of them. Which is clearly demonstrated by the smaller impeller and resulting lower MOI of it (another plus) providing a very similar flow potential.
.

Sorry, I didn't mean to take things out of context. But the 660, 7670, & 6465S do seem to flow about the same/similarly (60-65 lbs/min) on the cold side. The G30-660 apparently flows more than its wheel size indicates, which is an impressive feat! Tho I'm also curious how the 6465S would compare, especially with that "UHF" split-blade design on the hot side. Seems like the best of both worlds: lower MOI & backpressure while retaining the greater momentum capture of a high blade count. The turbine size is definitely closer to that of an 8374 tho. Correct me if I'm wrong but I always thought rotaries favored larger turbines (compared to an equivalent piston engine's requirements).

It would be good to see some actual results ..... often manufacturers promote something like this that sounds good in theory ..... but doesn't actually do much at all . They don't show turbine maps so without proper testing and comparisons ...who knows.

Bit of a generalisation. Rotaries do require bigger turbos for same power .. and large turbines work particularly well when there is a lot of overlap present , which is often the case with rotaries.

Mostly from the results of your build. An older generation turbo of a similar size (i.e. GT3071R) would have less flow and efficiency, thus resulting in less power. It's impressive how far Garrett has come wrt their blade designs. I bet one day we'll have compressors that can exceed 80% efficiency. ^.^

GT3071R: 53mm/71mm
https://www.garrettmotion.com/wp-con...p-GT-3071R.jpg

G30-660: 54mm/67mm
https://www.garrettmotion.com/wp-con...0-Comp-Map.jpg
Edit: Yikes! Those images are huge...is there a way to make them smaller?

Btw, what are your thoughts on trim size? I was comparing compressor maps of the EFR8374 vs EFR8474 which both have an exducer of 84mm but different inducer sizes (62mm vs 68mm), thus a trim of 55 (EFR8374) vs 65 (EFR8474). If we compare compressor maps of the two turbos, it seems to be trading efficiency for flow along with an increase in mass/MOI, of course. I bring this up because I noticed the XR6465S has a very low trim of 53.9 which makes me wonder how efficient it may be, in lieu of actual compressor maps. The low trim should at least aid in responsiveness though. But like you said, we simply won't know without actual results/testing.
https://cimg3.ibsrv.net/gimg/www.rx8...0997d10b12.png
Which WG are you going to use exactly? TiAl MV-R 44mm?

Lastly, back onto the topic of the G30-660... You went with the 1.01 turbine housing, but in hindsight would rather have gone with the 0.83? Is that conclusion simply due to availability or would you perhaps go with something along the lines of a 0.92 a/r if such a housing existed?

I misinterpreted your comment so deleted my response. Yes........ they have come a long way.

As I mentioned above ..low trim tends to give a map useful for a smaller engine running high boost so if you want the same flow more efficiently at lower boost (ie larger engine) you increase the trim. that is evident in the 8374vs8474 chart you posted. This is why I like the 660. Same is true for most of the G series.

I'm actually warming more and more to the 1.01 .... spoolup still very good and knowing that emap is well under control gives me the confidence to 'turn it up'.
Yeah ...if they made the 0.92 available for the G30 ... I think that would be my preference. The 0.83 is dropping a little too far IMO but it would certainly make spoolup exciting.

Thinking about Trim some more.
Two compressors of same design and same out put with different trim would suggest ...lower trim= higher Prs (small engine) ...higher trim =lower Prs (larger engine) . In this scenario the trim is different because of the different exducer diameter. Inducer would be the same.
But..................... if the trim is altered by increasing the inducer size rather than by decreasing the exducer ...... what we see is both more flow and at lower Pr. Per the 8374 vs 8474 chart above.

So in other words, a lower trim compressor would yield a more vertical map while a higher trim's map would be more horizontal? All else being equal of course...

But what I'm wondering is... what would be better for a Renesis that's aiming for a high redline? Say 8k+, even though most turbo builds tend to fall on their face by that point. This would suggest a high trim being preferable, yet the Renny is by no means a "large" engine. Would you consider 1 Bar to be high or low PR?

Yeah ..... that sounds like a good way to think about it .

Hey @Brettus, do you have a shaft speed sensor installed on your turbo by any chance?

No .... maybe one day I'll look at that . On a maf tuned setup you would only look at it a few times when setting max. power. After that you wouldn't need it as maf will tell you if you are leaking air to the point of overspeeding it .

So........... I find this really interesting (sad I know) :
Compare the Turbine housings :
G 30 vs GT 35
1.21 G30 is near to 1.06 GT35
1.01 G30 is near to 1.01 GT 35
0.83 G30 flows less than 0.83 GT 35
0.61 G30 flows less than 0.61 GT 35

The 1.01 is pretty much the same but all others flow less than the equivalent GT 35 ..... kinda weird.

https://cimg7.ibsrv.net/gimg/www.rx8...a233f52715.jpg

https://cimg5.ibsrv.net/gimg/www.rx8...9689a007db.jpg

it wan’t directed at you and imo you didn’t. Because as stated in a later post, there aren’t maps for that turbo that you can compare anything to.

the way it was challenged is what was being addressed.

if you want to get down into the details they aren’t “identical”, but the compressor/turbine pairing reference is still valid which is what you were referring to.
.

because it’s not just the housing, the wheel design influences it as well

imo this has been overlooked a lot and I never had that full understanding until studying the new G-series line. The turbo comparison thread over on RX7Club is a good example. All it takes into account is the average are of turbine inducer and exducer. If we did that with your G30-660 it’d look ridiculously undersized. Obviously it’s not.
.

which as previously proposed, the G25-660 0.92 appears to offer a better flow profile than the G30-660 0.83 with a number of other advantages for the ultimate spool-up


G25-660
https://cimg4.ibsrv.net/gimg/www.rx8...18322d960.jpeg


G30-660
https://cimg8.ibsrv.net/gimg/www.rx8...39fbcb245.jpeg

.

The Renesis (6 port) for turbo selection purposes should be considered as a 3.0L engine (approx.) ...so yeah ...pretty big. Small low trim wheels like the 7670 suit 1.8 -2.5 engines (or a 4 port Renesis). You should have a play around with the matchbot software as it helps understand all the factors that go into a system.

So what's stopping you from going for an 8k+ redline, if any? Is it just cuz you're falling off in the efficiency island (thus power as a result)? Fear of detonation perhaps? Or do you think you've hit the flow limits of the Renesis?

This: https://www.rx8club.com/series-i-maj...theory-268776/

Some will argue it's not relevant .......... I still think it is (on a stock Renesis anyway) until someone else wants to risk their engine to prove otherwise.

A rough overlay of the two maps; no doubt that the Garret compressor is a slightly better fit for either a 4P or 6P 1.3L rotary operating at the higher HP limit (efficiency centerline skewed toward the right compared to the 7670), but the turbine has to be factored into it as well. Which if the intention is to hold to that Renesis rpm limit theory, then again imo the G25 0.92 turbine combo needs to be seriously considered for most applications.



https://cimg9.ibsrv.net/gimg/www.rx8...e1595bb60.jpeg



Also stressing again that (1); Matchbot cannot accurately model the dynamics of a rotary engine. Moreover, (2); manipulating input parameters to force the results to match a known output is not accurate. As you can see in the more recent posts, even the Garrett turbine housings don’t flow the same between various models/generations. So right off the bat you have no idea how much different your non-BW turbine is from where the points are being placed on their turbine map or whether the emap and host of other features is an accurate presentation.

The design of the housings, impellers, and so on all matter. If you don’t have a BW turbo then you’re flying completely blind using that program. The differences between the G-series and BW as being discussed in this thread should make it obvious on just how different they are. That all matters if you don’t have a BW turbo, because the program isn’t accounting for those differences. It only accounts for their known turbo parameters.

.

Except, it's not too hard to take the flow numbers off the Garrett map and put them in a position that matches the flow on Matchbot (wasn't it you that pointed that out ?). So ................... totally doable.
Nice work on the comp. map BTW .

So with your current setup, what kind of flow rate would u say you're pushing by 7500rpm? And forgive my ignorance, but what's the flow rate of a 6-port N/A Renesis at 9k rpm? Also, you're in NZ... so your driver's side is on the right, correct?
Edit: I guess the answer is ~56 lbs/min?

Btw, I was trying to play around w/ BW's MatchBot thingy and was wondering what values you guys use to compensate for a rotary setup? Thx
Also, I noticed in your 2nd post that you had the 18psi line end at ~22psi (@7500rpm). Is that because you are expecting ~4psi of boost creep or something?
Edit: Nvm, I see that's probably what MatchBot outputted lol

Quote:

---

Originally Posted by Brettus (Post 4939907)

Except, it's not too hard to take the flow numbers off the Garrett map and put them in a position that matches the flow on Matchbot (wasn't it you that pointed that out ?). So ................... totally doable

---

I understand why you think it should be that simple, but perhaps it only signifies that the message about Matchbot being specifically programmed around the parameters of the BW turbo characteristics is still not registering. Because the statement quoted above is based on multiple assumptions ignoring the fact the G-series is setting a new standard in turbocharger performance. In addition to also ignoring that a program designed around reciprocating engine parameters is not going to accurately model the characteristics of a rotary engine. Therefore any and all outputs are highly suspect, if not entirely inaccurate.
.

What you don't seem to understand is that almost half of the values inputted are guesses anyway. So the accuracy of the outputs are only as good as that of the inputs. By actually measuring a lot of it and through experience you can eliminate most of the inaccuracy and get some good data. That is the same whether it's a piston engine or a rotary. You seem to have it in your head that there is some kind of majic sauce that skews the result for a rotary.
You have differences in BSFC, exhaust temp,Ve, and with a two rotor twin scroll , turbine efficiency, . Those parameters can be quite a bit different on a rotary ...but if you can get a handle on them ...it's just maths.

Here is one I did for my setup. The specific things you need to watch for on a rotary are :
BSFC
Exhaust gas inlet temp
Volumetric efficiency
Turbine efficiency (twin scroll PP engines can have great efficiency early in rpm range)

https://cimg3.ibsrv.net/gimg/www.rx8...a8d67af938.png

And yes stg is on the right here in NZ ...why do you ask?

Ah, I was just wondering... how much more space would you have had down there if your RX8 was LHD instead of RHD? Is the difference minimal? I'm basically hoping that w/ a LHD build, there'd be more wiggle room. But by how much? You've packed everything in there pretty tight, kudos lol! I'm still only halfway through reading your "Ultimate Renesis" thread (it's 94 pages long!)... XD

LHD def. has an advantage. It's quite a mission getting around the steering column with the pipe to the IC. Otherwise..................... much the same.

Btw, I'm curious... have you considered direct injection? Not just DI by itself, but rather, in conjunction with the existing port injection. I was sorta inspired by Toyota's D4-S system where they use both port and direct injection simultaneously. I'd probably want to place the direct injectors in the same place Mazda did on their 16X prototype motor, i.e. directly on top & pointing straight down. I have a totaled BRZ so I was thinking of harvesting the direct injectors from the FA20 motor. Anyway, technical implementation aside, what are your thoughts? Do you think this merits some investigation? Or would such an effort be more trouble than it's worth? I just thought this seemed like the "experimental" type of thing you might attempt lol.:yelrotflm

https://www.rx7club.com/attachments/...tinjection.jpg
https://www.rx7club.com/attachments/...-injectors.jpg

Edit: Oh and, the spray pattern of the FA20's direct injectors are shaped in a dual-fan pattern:
But of course, there are various spray patterns from other vehicles as well... (e.g. multi-point, cone-type, etc...)
I imagine with the given angle, the fans would hit the sides of the "bathtub" of the rotor and deflect towards the centerline, thus swirling perpendicular to the rotor's axis of rotation.
http://www.ft86club.com/forums/attac...1&d=1326102800

Ah, 1 more thing... have you tried WPC treatment before? If I remember correctly, I think Mad Mike uses it on all of his builds. It supposedly works wonders on rotaries due to all the frictional surfaces.
P.S. I love all the R&D you're doing on the Renesis, and I pray for your success!

Cheers staf
Direct injection : haven't thought about it , that would be an entirely new project . Possibly bigger than the turbo project itself.
WPC : don't know a lot about it .... what are the gains ? NVM ...watched the video ...

It's basically a surface treatment (not coating) that reduces friction like no other. It actually goes well w/ the existing nitriding finish or other coatings as well. It's more of a reliability mod than anything though... as in, I wouldn't expect any power gains. Although, if there were power gains to be had (due to less friction), it would be in the single-digit range. I imagine if your previous motor(s) had WPC treatment, then these results you posted in your other thread may have been avoided or at least minimized:
https://cimg1.ibsrv.net/gimg/www.rx8...fa8e6719bd.jpghttps://cimg1.ibsrv.net/gimg/www.rx8...ec200a54bb.jpg

Ok, forgive my continued pestering... I am rather late to the party and have much to catch up on lol. Anyway, I was wondering if you're still using the intercooler featured in your "Ultimate Renesis" thread (I finally finished all 94 pages!) and if so, do you know the dimensions of the intercooler core? Also, did you specifically pick that type of intercooler design for the vertical end-tank layout? (i.e. as opposed to having the end-tanks on the left/right ends) This is the one I'm referring to:
https://cimg7.ibsrv.net/gimg/www.rx8...6253e6e975.jpg
Just asking because I'm trying to figure out if this intercooler would fit in an RX8. Even if it fits, I'd love to discuss the pros/cons of the differing end-tank styles. Thx!

The one in the link looks too small for anything above 300whp ...you haven't stated any goals so :dunno:
Mine may be similar core size but because all the tubes are short and there are lots of them .... it is more efficient and has very low pressure drop.
I could benefit from a larger core for things like dyno pulls because in that situation you need a lot of mass to hold temp down without much airflow over the fins. But for the track etc it works great.

I c... well, the Mishimoto one I linked is rated for 500HP supposedly, but that could just be an optimistic figure since it's coming from the manufacturer. I don't really have a concrete goal other than to further Renesis R&D lol. But if I had to pick a HP target, then I'd say 400hp tops for a semi-realistic goal. Probably using that XR6465S turbo I mentioned before. I feel it's always nice to have more data for the community, hence trying stuff others have not.

However, if I had an unlimited budget, then I'd say my dream build would be a short-shaft 3 rotor Renesis using 4-port front/rear irons and 2x 6-port center irons (i.e. the reverse of what you did). I remember seeing a 3-rotor Renesis done by Hayward Rotary in the UK (I think?). Would probably utilize REC's CNC porting service for consistency as well (especially exhaust ports). For FI, I might go w/ a Rotrex C38R to accentuate the N/A Renesis powerband and rev to the moon. Yes, the middle rotor would be choked to some degree, but I wouldn't be trying to break any HP records or anything. If that was the goal, I'd go turbo instead, but I might as well go PP 20B if I did that (as opposed to a "20B-MSP"). It would mostly be for the "feel" and novelty I guess. Oh, and for science!! I'm guessing such a build would sit around the 500-550HP range. A guy can dream, right? lol

Honestly, that 3-rotor was my "long-term" plan for my BRZ before it was totaled, but since that's no longer an option I've decided to go back to my roots and purchase another RX8 and go from there. (I learned to drive stick in an RX8 long ago, in a galaxy far far away) Obviously, a 3-rotor is still a far into the future prospect, so my most likely course of action will be to slap a turbo onto a stock RX8. I'll probably pull the trigger within the next couple of weeks after I've settled into my new house (man, moving is a bitch). Anywho, I digress... but those are my "goals" more or less (sry for the tangent). :smoker:

Edit: Btw, I'm not sure how thick your IC core is, but that Mishimoto one is a hearty 3.75" thick. So maybe that's why it's rated higher than it looks. Fwiw, it was tested on a Hyundai Genesis @ 576rwhp.

Yeah ...500 piston fwhp = 450 piston whp = 340rotary whp .... So pretty good rating considering how big it is (must be efficient design) ..... but still too small.

3 rotor : https://www.rx8club.com/series-i-maj...2/#post4930494

Hey Brettus, I was just watching this video on the new Supra and its B58 engine. I noticed the way BMW designed the exhaust manifold is kind of similar to the design of your manifold, at least functionally; except their's is integrated into the block. Basically no runners, with a small volume from the exhaust to turbo. I guess BMW concluded that pushing 3.0L worth of exhaust through those 2 fairly small exhaust ports (see video @ 3:55) wasn't much of a bottleneck. That just reassures me even more wrt your manifold design! I know it's an apples or oranges comparison, but still lol. Edit: I'm assuming the inside of that integrated manifold is hollow, but it may not be now that I think about it...:dunno:
Edit2: Turns out they decided to make a more traditional exhaust manifold, so idk what to think now lol. (I didn't know you could 3D print Inconel!)
Man, I would love to see your manifold made using this process!:yelrotflm

It's been modelled and is ready to 3d print and test fit . Then we will investment cast a trial version in stainless.
It's gunna look better, flow better and boost creep less than my fabricated one for sure :)
But as far as flow goes ..... There is close to zero restriction at either end under full load and that only leaves the siamese being slightly restricted which with all the testing I've done ...just isn't an issue.

Brett,
Any idea on the timeline for you getting the new manifold? How much do you want for the used manifold after you get the new one? Maybe just have several made at the same time?

I think I could sell my existing one 10 times over ...lol ... But I'm not happy with level of boost creep on it as it is anyway ... most of that is due to the extreme efficiency of the G30 though !
Should have my 3d print ready for test fitting this coming week ...fingers xd.
Once I get the actual casting there is still a lot I need to do to make sure it works as intended. Maybe some track testing etc . I really don't want to be sending something out I'm not 100% confident in!

oohhhh the 100% requirement - that's sad news. I have been following your turbo research for years and it doesn't look like you will ever stop looking for improvements which means you won't meet that 100%. Unhappy

Sorry for the highjack, regarding intercoolers, just replaced my Greddy one with this https://www.holley.com/products/cool...rs/parts/FB604
Ducted, about 17C rise from ambient temperature. Drop in replacement for the Greddy, no front bar mods needed. It's partially ducted (I need to seal the underside). Recommend it 100%.

.... would it make you feel better if I said it's 99% there already ..... :lol:

Is it a way to generate "read protected" 3D files?

I'm thinking that if this is possible, you could sell the files for financing the last 1%:), no/little hassle for you. Metal 3D printer availability seems to be on the rise, and I remember Koenigsegg mention that 3D printing was more cost efficient than cast, and revealed new possibilities for them. Now this was some years ago, casting cost might have changed, and the possibilities might not be needed to a manifold, but still like the idea of just sending a file that we can print.

Sounds very cool ....

What material will you be using for the 3D print? Inconel by any chance? :p

It will be printed in a PLA material for the lost PLA casting process. Then cast in stainless.

Brettus, these guys are in your neck of the woods and can give you a quote for metal 3d printing (stainless 316, titanium, etc.) to compare against your above PLA method. https://www.ram3d.co.nz/

Ah, so you're 3D printing a mold, not the part itself. That does sound more cost-effective.

I'm just going with what my engineer is recommending ..... It's been 16 years since I was an engineer so I've completely lost touch with the best way of making things these days.

I'm still a practicing engineer and I wouldn't trust me in that field! :lol:

First test of 3d printing into nylon done . It has a few blemishes from the process but otherwise looking pretty good !


https://cimg4.ibsrv.net/gimg/www.rx8...c1f3709fcd.jpg
https://cimg8.ibsrv.net/gimg/www.rx8...a58b20955b.jpg