Hydra uses a mhi wheel though so they may have compressor maps available. He says the ratings are at peak efficiency not peak output. It's similar to our stock turbos are able to push 25lb but factory rating is much lower
2019 upgraded twins choice for 650whp
- Thread starter Bmxr152
- Start date
I had MMP 3's and had to rebuild twice in a year and i got them with zero miles 'rebuilt'...really makes me wish i had gone with pures or a budget single.
I wouldn't expect a wheel alone to be indicative of flow. I could be misunderstanding, but I thought turbo flow was dependent upon the entire package, turbine wheel, housing, a/r, and compressor wheel, housing and a/r.
Wouldn't you have to test a turbo to create the compressor map?
Exactly. I don't see how you can do that without a real compressor map. Which you can't have for custom turbos with mis-matched wheels, housing, etc.
This video explains it well:
It's all a guesstimate without a compressor map.
Here is a good list of turbo lb/min: https://www.evolutionm.net/forums/evo-x-dyno-results/692309-turbo-flow-rates.html
MHI (Mitsubishi) OEM Evo X turbos:
Turbo Part No. Flow Rate (lb/min)
Stock Ralliart turbo TD04HL-15T-7 33
Stock Evo 10 turbo TD05H-152G6-12T 39 (reports vary from 36-42 – this is my best guess)
Upgraded "MHI 18K" Evo 10 turbo TF06-18K or TF06-07*18KX3RC-12T 54+
Forced Performance (FP) Evo X-specific turbos:
Turbo Flow Rate (lb/min)
FP 71HTA 51
FP Green 54
FP Red 60
FP Black 68
CBRD Evo X-specific turbos:
Turbo Flow Rate (lb/min)
BBX Lite 51
BBX Full 56
RAD Lite (Ralliart) 36
Approximate numbers for the EF series from another forum
EF2 ~51lbs/min
EF2.5 ~54lbs/min
EF3 ~61lbs/min
EF4 ~72lbs/min
Popular universal aftermarket turbos:
Manufacturer Turbo Flow Rate (lb/min)
BorgWarner EFR 6258 44
Garrett GT3071R 47
BorgWarner EFR 6758 49
Garrett GT3076R 52
BorgWarner EFR 7064 56
Garrett GTX3071R 57
BorgWarner S256 58
Precision (PTE) 5857 60
Precision (PTE) 5858 CEA. 60+
BorgWarner EFR 7670 64
Garrett GTX3076R 65
Garrett GT3582R (AKA GT35R) 65
Garrett GTX3576R 65+
BorgWarner S362 65+
Forced Performance (FP)
HTA GT3582R (82HTA wheel) 70
Precision (PTE) 6262 74
Precision (PTE) 6262 CEA 74+
Precision (PTE) 6265 74+
Precision (PTE) 6266 CEA 74+
Garrett GTX3582R 75
Garrett GT4088R (AKA GT40R) 75
Forced Performance (FP)
HTA GT3586R (86HTA wheel) 75+
BorgWarner EFR 8374 79
Garrett GT4094R 80
Garrett GT4294R (AKA GT42R) 84
Forced Performance (FP)
HTA GT3794R (94HTA wheel) 90
Precision (PTE) 6466 CEA 90
Precision (PTE) 6765 90+
Precision (PTE) 6766 CEA 93
BorgWarner EFR 9180 94
Garrett GT4202R (AKA GT42RS) 95
1 lb/min of air flow = roughly 10 HP. Hence a 6466 turbo being rated or known as a 900HP turbo, 90 lb/min x 10 = 900.
How much better is this 9180 turbo map?
Last edited:
Yes, my car is rwd. I am running Mickey Thompson ET Street S/S 275/40 R17 tires with 3.25 diff ratio. Suspension is nothing special - Eibach springs, Koni Yellow shockies, HR stabilizer and Nolathane subframe bushes.
Most small turbo shops (i.e. not Garret, BW, etc.) do not have the capability of compressor mapping without farming it out. I've done it before, it's actually very cool to do, but involves some very high dollar equipment. I have been kicking around some ideas on how to simplify this -the reducing cost of instrumentation is slowly making a setup to test/measure this be more feasible, but as of today it's still big bucks.
You can backwards calculate an effective compressor map using a dyno and a lot of time, but it's nothing like the real thing. Conversely a real compressor map doesn't necessarily tell you what HP you're going to make, as that's a function of the whole system, it's just a rough guide to approx. capability.
The N54 is good for making healthy HP for a given flow, the 10hp/lb is a good rough estimate. Anything you can do to drop that pressure ratio can often really help, as is easily visible in damn near all compressor maps. I.e. your plumbing, intercooler, charge pipe, manifold, head, etc all matter. Bringing us back around to "an engine is an air pump".
Chris
You can backwards calculate an effective compressor map using a dyno and a lot of time, but it's nothing like the real thing. Conversely a real compressor map doesn't necessarily tell you what HP you're going to make, as that's a function of the whole system, it's just a rough guide to approx. capability.
The N54 is good for making healthy HP for a given flow, the 10hp/lb is a good rough estimate. Anything you can do to drop that pressure ratio can often really help, as is easily visible in damn near all compressor maps. I.e. your plumbing, intercooler, charge pipe, manifold, head, etc all matter. Bringing us back around to "an engine is an air pump".
Chris
Most small turbo shops (i.e. not Garret, BW, etc.) do not have the capability of compressor mapping without farming it out. I've done it before, it's actually very cool to do, but involves some very high dollar equipment. I have been kicking around some ideas on how to simplify this -the reducing cost of instrumentation is slowly making a setup to test/measure this be more feasible, but as of today it's still big bucks.
You can backwards calculate an effective compressor map using a dyno and a lot of time, but it's nothing like the real thing. Conversely a real compressor map doesn't necessarily tell you what HP you're going to make, as that's a function of the whole system, it's just a rough guide to approx. capability.
The N54 is good for making healthy HP for a given flow, the 10hp/lb is a good rough estimate. Anything you can do to drop that pressure ratio can often really help, as is easily visible in damn near all compressor maps. I.e. your plumbing, intercooler, charge pipe, manifold, head, etc all matter. Bringing us back around to "an engine is an air pump".
Chris
Tony, you posted under Chris again.
JK I am sure Chris has made compressor maps, probably at his day job.
Perhaps you can share some of the ones you made?
What turbos did you make them for and why go through the hassle?
What would you guesstimate GCs flow lb/min?
We agree on one thing, an engine is an air pump!
So I'm pretty sure I'm going with the speedtech kit. Now the only issue I see with it is killing o2 sensors constantly. Anyone have any experience with the 6266 spool times? My evo with a 6266 made 31lbs of boost by 5.2k rpm lol. Also anyone with the speedtech kit get the manifold with ceramic coating? Is it an actual legit coating or some cheap coating that comes off after a while?
I think @veer90 has a 6266 and had mmps before that. I recall doing an overlap of his dyno numbers and the 6266 spooled better in a steady State single gear acceleration.
Transient response on the street might not be the same, but I'm sure since he has first hand experience with both examples he can better explain the difference.
I hit 24psi by 3900rpm with a speedtech kit and an s257 with a .88. the ADVs Jeff sells also have been working since install.
Compressor maps made were for turbine engines, proprietary stuff that I can't share, but I can say the engines I've done testing for are on many planes that you've probably had flights on (Boeing, some of the big Airbus) and more often they're on planes that you probably haven't had rides on (but maybe) such as the nice Gulfstream business jets, and some that I hope you haven't had to have rides on (secret). In this case it's important to be able to define the flight envelope for starting, operability, and performance (how high, hot hot, how cold, how much load, what type of load, etc). It's extremely similar to turbos, which is why there is so much transference between the two careers. In fact, take your compressor output, feed it into the turbine inlet, add some fuel and a flame source and you'll make your own turbine engine. Few youtube vids of some brave individuals doing such a thing.
Some of these guys have no clue how close death is. *shudder*
Mapping is a large scale effort, often testing is accomplished in an environmental chamber that simultaneously controls air temp, conditioning, and density for the chamber independent of what we're feeding the motor (allows us to replicate weird scenarios such as being up at 40kft, having everything soaked to -65F for hours, then needing to start during the descent or landing on a hot day, so motor is cold soaked, altitude is whatever we want, inlet temperature is hot day). Standard engineering approach, stack the odds in every which way until you've tested every conceivable possible iteration that is possible in flight. The compressors we use for the test chamber control are beyond enormous, 3500 horsepower or so, which is so large in fact that we're prohibited from running full scale during the day, as the electrical demands are so significant. It's expensive too, the better part of 20k/day to run one of these tests, there is no such thing as a test that's less than 6 digits and often pushes well into 7.
Why do a compressor map for a turbo? It's a good question. I'm not sure it's worth the hassle for what we're doing here, although it would allow you to no-shit understand the flow differences between blade iterations and you'd see the edge (surge or stall) long before you got there, as well as understand how that particular design behaves near that edge. Our flow and performance guys are among the best on this earth and they'll freely admit no computer program (yet) is capable of analyzing all the data such that we can do it via analysis vs. test, which is one of the reasons I love testing so much, and also a reason I've enjoyed working with Tony. You're not going to find him doing CFD on a new blade design, but he's so aggressive with procuring, testing, and categorizing performance that I am impressed, and at the end of the day how she performs on the street is what matters to most of us. Anyway, we make these maps to understand both incremental and significant changes. The aviation industry often qualifies parts through similarity, but it can become a runaway train. Minor tweak to a compressor, no re-qualification testing needed. Minor combustor tweak, no test needed. Minor turbine blade tweak for manufacturability, we should be good there everyone agrees. Oh, fuel control vendor went tits up, let's cost reduce her a bit, qualify it on a component level and we're good. 10 years later you've really lost your footing on what the limits are, and when cert requirements change, and customers don't want it qualified to the old standard you've been claiming similarity to, at some point you have to own up that you prob need to revisit all the stuff, if for nothing else than making models that reflect real-world performance to help further design iterations. I've wandered a little bit, this field is something I spend most of my time working on. Bottom line is a compressor map shows the capability of that particular compressor, and will help you avoid some issues once in service.
Arguments against it are cost & effort first & foremost. Additionally a compressor map is still completely de-coupled from the system it's in. I.e. that's the compressor.... with what turbine? In what housing? At what temperature, altitude, etc? Once in the car, with what inlets, what outlets, FMIC, charge pipe, throttle, head, valves, etc? All of this impacts the delta P across the compressor (among a million other things), so I tend to view compressor maps as more a tool of comparing compressor to compressor (i.e. advanced blade designs) versus suitability for a particular application -too many other variables in turbo design that will make it sink or swim.
Once you have some experience goofing around on a motor, you can glean rough power potential from a compressor map, if you assume the rest of the system isn't completely stupidly designed (usually.... but not always). The N54 is efficient at turning flow into whp. An estimate on GC v2.0 capability would just be backwards calculated from our testing, which would lead you to somewhere around 40lb/min, but this is just shit kicking talk, nothing authoritative. Tony actually may have a better idea. I'll say this, there are PLENTY of gains to be had by flow-optimizing the system; all of us are paying the penalty of envelope compromises. I.e. the stuff has gotta fit in the space we have, optimized flow be damned.
That's enough wandering for a Sunday morning. I'm going to have more coffee and chase the wife around.
Chris
Yes, scary is correct. Turbine engines have containment criteria, and since I'm planning a containment test right now I can tell you that the criteria is that the blades can go through the motor, outside of the motor, can dent or even create a small tear/hole in the metal shield a foot or two outside and around the motor (think skin of airplane), but no material can fully penetrate that outside skin. In other words, stuff that is designed to be contained in the event it blows up still would completely fuck you up if it let go, if not kill you. Stuff that is NOT designed for containment (read that: 99.9% of turbos) it's up to the probability gods on if a piece hits you or not. You likely would not survive and it would be a mess if a hub burst occured. I've been in cells plenty of times next to motors that are running safe, measuring hundreds of parameters with safety shutdowns, watchdogs, even heart beat monitors (instrumentation heart beat to ensure your data is good). Overspeed is 100% shutdown worthy no matter what, even in essential mode. These guys aren't monitoring any of that stuff. I wouldn't hang out next to one of those home made motors being pushed for any amount of money. In fact, be wary of such a thing next to a dyno IMO too. I'm not one of those crazy safety guys, there is just so much energy involved one ought to pay proper respects. I like pushing limits and goofing around the same as anyone else, but having an 800 F compressor spinning at 100k RPM rip through me isn't high on my list.
Possibly I can post carnage pics once the test is done, but that won't be for months.
Chris
I'm surprised to see the amount of dislike and lack of recommendation for MMP's turbos honestly. I know the story of the engine that Bryan whateverthefuck got smeared their reputation a bit because he is a mod on most N54 FB groups, but I've always thought it was obvious they were the most forward-thinking and well engineered parts for performance, especially turbos. I don't think any other twin vendor has ever offered coated or ball bearing twins, and until recently they were the only ones who had custom manifolds - and despite what Rob Beck would say about them, they were clearly designed with a lot of thought put into them. Also pretty much anyone who's serious about 1/4 mile times and on twins runs either of the two top tier MMP turbos judging from what I see in the private drag racing-oriented FB groups. I guess the sample size as well as end-goal uses are different here and there by quite a bit.
I think the main reason you see "a ton of" rebuilt MMP S3 or 1Ks for sale is because people expect to achieve that 700+whp high-boost on e85 blend numbers right off the bat without having to add all the supporting mods and fueling, and then realize it will be easier to make their lower/realistic HP goal on 93 with a ST. And then MMP offers to transfer their warranty if the seller sends the turbos in for inspection and if needed rebuilding but calls them all "rebuilt" rather then separate groups like "inspected and golden" or "we replaced the scuffed up V band clamps".
Personally I've loved my MMP 1Ks and will always recommend them to anyone seeking over 600whp on twins. Performance wise they're amazing and pull easily until redline (and then some actually, considering I rev past 7k now) and spool time is definitely slow or laggy, with the proper supporting hardware and the right tune you can still achieve impressive spool times especially for turbos that are so much larger than stock (quite a bit larger than S55 turbos too iirc?). Not to mention I had excellent customer service from Mauricio too both before, during and after the purchase of my turbos.
I think the main reason you see "a ton of" rebuilt MMP S3 or 1Ks for sale is because people expect to achieve that 700+whp high-boost on e85 blend numbers right off the bat without having to add all the supporting mods and fueling, and then realize it will be easier to make their lower/realistic HP goal on 93 with a ST. And then MMP offers to transfer their warranty if the seller sends the turbos in for inspection and if needed rebuilding but calls them all "rebuilt" rather then separate groups like "inspected and golden" or "we replaced the scuffed up V band clamps".
Personally I've loved my MMP 1Ks and will always recommend them to anyone seeking over 600whp on twins. Performance wise they're amazing and pull easily until redline (and then some actually, considering I rev past 7k now) and spool time is definitely slow or laggy, with the proper supporting hardware and the right tune you can still achieve impressive spool times especially for turbos that are so much larger than stock (quite a bit larger than S55 turbos too iirc?). Not to mention I had excellent customer service from Mauricio too both before, during and after the purchase of my turbos.
Last edited:
I think reg stage 3s are not bad but the 1k if you ask anyone that designs or knows turbos will tell you it's too large for the n54 head. I think mumps issue is he may push little ti hard to fast and come to market to quick si he had inlet issues and he got dragged into pissing contest with rob about the motors which he then ran his mouth and got caught in some exaggeration. The only thing I didn't like was the fake injector thing but I agree I dont think his turbos are bad at all and are a decent high power setup. It's just for price pure has better track record and now rb gf have the same power potential with better design , it's just the market changed and their are better options imo ,that doesnt mean they are bad or a mistake
I wonder who on this platform, especially the guys coming from other turbo platforms, would say that the N54 head CFM is designed in any way to effectively make horsepower over any number well below numbers being imagined here (ala S3/1K/etc)? Head work is common in most performance situations I know of, yet I was surprised when I came to the BMW turbo world to find that near no one was doing it - just cranking of the boost and arguing the merits of various boost crankers
.My uneducated observation is that many people on this platform look for high, sustained boost as the answer, and whatever breaks first is the problem.
Filippo
@fmorelli Quite a lot of people that are easily impressed by big numbers, especially boost. How many times have you seen "I pushed my stocks to 26 psi"? Or something similar... Way off the efficiency island at that point and down on performance compared to 17-21 psi given the metric ton of hot air being shoved around.
Look at jb4 ,when you do backend flash boost drops quite a bit cause it's the better choice with stock turbos to run leas boost and more timing.
And @filippo yes to your statement but the 1k Is worse then normal cause of extreme mismatch in runners. With turbos you want the heat and velocity high as a mofo with least restriction until that turbine
The dislike and lack of recommendation is from the high failure rate. I know 3 locals that had them and all 3 had numerous issues ranging from complete failure, smoking out of the box, wastegates falling off, and on an on. Two of the three also lost motors shortly after installing the gen 1 port injection with "genuine bosch" aka Chinese injectors. In the beginning I think lots of failures were kept quiet by given store credit, discount parts etc. The reason you see them all fresh rebuild ones for sale is because most have failed under warranty. I think the nail in the coffin was the 1ks. Extremely laggy with not much increase in performance over the stock manifold stage 3's.
