Announcement

Collapse
No announcement yet.

Turbo choices for M20

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Turbo choices for M20

    Slowly building my turbo m20 and curious about turbo choices. I want a quick spool/250-300hp. Back in the day I had a T3T04E on my 4cyl SR20DE, lots of lag, what I want to avoid.

    So I keep coming back to the Garrett GT3076R .63A/R. Seems like you get strong boost by 2500 rpm. But the price is so dang high. The HX35 is cheaper but from what I can find it hits around 3500rpm. Are there other good options? I see a ton of used Subaru turbos for sale, any of those? Not too comfortable buying a Chinese ebay turbo unless you can talk me into it.
    '90 325is

    #2
    A GT3076R is way overkill for 250-300hp (see below). The T3/T04E is cheap for a reason, it's a very old design so it definitely doesn't spool as well as modern turbos with improved aero, but your M20 has at least 25% more displacement than the SR20. 250-300hp is super easy, any turbo that will make that without huge overhead will be plenty responsive.

    My turbonetics 60 trim T3/T04E will do maybe 450hp worth of flow, probably less with the .63AR exhaust housing and internal wastegate I have. It starts making boost reasonably quick around 3,000rpm, rolling on in higher gears spool to full boost is occurs in a second by 4,000rpm, and full boost is irrelevant through 2nd since it spins the tires at anything over 13ish psi and revs out very quickly even with my 3.25 rear end. If I floor it at 2,000rpm I'll see positive pressure in a second or so. I previously had a 50 trim with the same turbine, it is the same story but it's a 400hp or so turbo.

    T3 super 60s are the biggest of the non-hybrid T3s and will be super responsive; substantially smaller compressor than a T04E, my inducer is nearly as large as a T3 super 60 exducer. Anything in the T3 size class will work for your modest goals and be quite responsive. TD04 16T, T3 60 trim or super 60, GT2860R/RS, Precision 48XX/5XXX, BW S250 or SX-E? Just pair it with an appropriate turbine and housing.

    Click image for larger version  Name:	GT3076R.jpg Views:	0 Size:	138.2 KB ID:	10087101
    GT3076R compressor map


    Click image for larger version  Name:	T04E 60 trim.gif Views:	0 Size:	13.0 KB ID:	10087100
    T04E 60 trim


    Click image for larger version  Name:	T3 Super 60.gif Views:	0 Size:	8.9 KB ID:	10087102
    T3 super 60
    Attached Files

    IG @turbovarg
    '91 318is, M20 turbo
    [CoTM: 4-18]
    '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
    - updated 1-26

    Comment


      #3
      M20s seem to be hard to perfectly size a turbo for. I have given much thought to it, and it seems that, for the most part, when you size a turbo for 250-300hp, it creates a lot of heat up top. You can correct that with an exhaust housing, but then you hurt the spool. And then, in the other direction, the s256 is commonly used, but is lazy.

      The sweet spot, in my humble opinion, is the g25-550/660 stuff, copies made by Pulsar. They used to be a cheap company, but have really stepped their game up and are making high quality copies of the fancy new g25 and EFR technologies, but for less than half the price. Dual Ball bearing so you get the transient response you don't get from the journal bearing affordable borg warner stuff.

      I have an e28goodies manifold, so I really wanted to stay twin scroll (and also have a 2.7L), so I opted for the Pulsar g30-660, simply because they had a twin scroll exhaust housing available on the dealer side that's not available on the retail side yet. There isn't a divided exhaust housing available for the g25s unless you go divided vband.
      1990 Brilliantrot 325iS Build Thread
      1989 Zinnoberrot M3 Build Thread

      Comment


        #4
        Thanks for all the compressor maps Varg. Tell me if I am reading them correctly though.

        I went to a turbo calc and entered the following:
        RPM: 6000
        Engine size: 2494
        Vol Efficiency: 85%
        Boost Pressure (bar): 0.7

        So 300hp at 6k

        It gave me a ratio of 1.7 and airflow of 29 lbs per min.

        Using this, and looking at the T3 Super 60 wouldn't that be putting me out on the choke line? For the T04E 60 trim and GT3076R that's right in the sweet spot. Or is my Volumetric efficiency wrong (I just left the default)

        ​
        '90 325is

        Comment


          #5
          Those numbers seem reasonable enough, in the ballpark at least. Higher VE will lower the boost pressure for a given power output, your cam and turbine will affect this too. I think you'll need more boost than 0.7bar for 300hp (~260-270whp) with a T3, the turbine will hold you back with a 2.5. You may want to find a copy of Maximum Boost, it's out of date on some things these days but the basics remain, including turbo sizing.

          That's also the smallest turbo I listed, and is going to be tapped out at 300hp, so you'll be making that power at a higher pressure ratio as you stray from the peak efficiency range and have a small turbine holding you back, but it'll spool early and fast. T3s spool fast on 2.1-2.3L Volvo redblocks I have run them on in the past, they'll spool super fast on a 2.5L M20 which has a much better head. If you want 300whp you're probably going to want something bigger. The turbo choice boils down to more than the compressor, you can tune your EBR and power band with turbine and turbine housing sizes. I ran a .48ar housing on the aforementioned T3/T04E 50 trim when I had my M42 and it was plenty responsive despite the low compression ratio. The 60 trim with a .63ar housing on the M20 is perfectly sized to make my turbo M20 responsive enough that I don't have to downshift all the time and still have a traction limited car capable of low 12s on high boost.​ It's basically the same turbo with a different turbine housing being used on a 1.8L and a 2.5L successfully. The compressor difference is pretty small.

          Originally posted by AWDBOB View Post
          M20s seem to be hard to perfectly size a turbo for. I have given much thought to it, and it seems that, for the most part, when you size a turbo for 250-300hp, it creates a lot of heat up top. You can correct that with an exhaust housing, but then you hurt the spool. And then, in the other direction, the s256 is commonly used, but is lazy.
          That's every turbo engine and every power goal. You are always compromising between maxing out a turbo setup to make the power you want but having good response or having worse response but the same power on less boost with lower egt. That's why VGT and twin turbo setups exist, and it's why even older factory turbo cars which can be beat on safely at OEM configuration have all of that tunable overhead, but will fail if you beat on them with the spicy tune that maxes out the stock externals.

          IG @turbovarg
          '91 318is, M20 turbo
          [CoTM: 4-18]
          '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
          - updated 1-26

          Comment


            #6


            Originally posted by varg View Post
            I think you'll need more boost than 0.7bar for 300hp (~260-270whp) with a T3, the turbine will hold you back with a 2.5.
            How much boost can you safely run on an M20 with a stock head gasket and arp studs?

            You may want to find a copy of Maximum Boost, it's out of date on some things these days but the basics remain, including turbo sizing.
            I have it 😁 when I turboed my sr20 in 1999 I bought it but didn't read it. No time like the present, plus I'm a little wiser now. Only a little.


            The 60 trim with a .63ar housing on the M20 is perfectly sized
            This is what I am thinking. I want small to get a quick spool but not so small I limit the potential. I know once I'm back in the boost I'll want more.

            Editied...
            Choices:
            1. The pulsar 4849 AWDBOB mentioned looks interesting too, map looks good and nice price.
            2. td04-16t or 19 from volvos and Saabs
            3. The VF39's on 02-06 STIs look like a TD04 with 60mm trim on the compressor from what I can tell. And there are loads of them on facebook marrketplace. (46.5mm/60.071mm)^2 x100= 59.92mm trim) now not sure how the exhaust side compares.
            4. Or just get a brand new T3 T04E, 60mm trim and .63A/R with journal bearings for $800. The link is to a local place so I'm leaning this way. I can go there and see if they have a deal for me on a ball bearing version :)
            Also found this on how much you can boost the M20:

            Originally posted by kamotors
            yeah here is my new out look on the M20.

            0-10psi=stock headbolts,can get away with WAR or other tuning methods
            10-16psi=ARP head bolts+MS
            17-20psi=+wasted spark
            20-23psi=+welded coolant channels
            23+psi=Build the snot out of it- ARPs,welded head,oring block and beefy lower end.
            ​

            Since I have MS, wasted spark, and plan ARPs it looks like I can easily run 14psi so I will review my compressor maps based on that :)
            Last edited by Hoffbrohaus; 03-09-2023, 01:42 PM.
            '90 325is

            Comment


              #7
              With those modest goals I'd probably get a Precision PTE of similar size to the T3/T04E 50 trim and a .48ar housing or a GT2871R or similar BB PTE depending on budget. A Pulsar may well be a much better choice, I just have zero experience with them and haven't researched or heard much about them. I do know that a T3/T04E 50 trim is very efficient at 26-28lb/hr and a pressure ratio of ~1.8, and that full boost by 3,500rpm with a rev limit of 6,700rpm is a good deal.

              I don't really like take-off turbos like VF39s and late Volvo turbos because the weird flanges introduce more complexity and fab work to the setup. To my knowledge you can't buy a manifold that will bolt a VF39 to your M20 and you need to buy a goofy exhaust flange or parts of the subaru downpipe to make an exhaust for it. I don't know anything about pulsars, unfortunately. I also don't like how full-race doesn't list the turbine size for that garrett. See below:

              Originally posted by A reference doc I've had for a long time
              T3 Turbines
              "Stage 1" - inducer: 2.320; exducer: 1.916; trim: 68
              "Stage 2" - inducer: 2.555; exducer: 2.120; trim: 69
              "Stage 3" - inducer: 2.555; exducer: 2.227; trim: 76; aka "T31"
              "Stage 4" - inducer: 2.795; exducer: 2.437; trim: 76
              "Stage 5" - inducer: 2.798"; exducer: 2.439"; trim: 76; aka "T350" (this may not be correct?)

              Turbonetics F1 turbine wheels (some of these seem to overlap in dimension with the T3 wheels listed above)
              F1-49 - exducer:1.929; inducer: 2.320; trim: 69
              F1-54 - exducer: 2.126; inducer: 2.555; trim: 69
              F1-57 - exducer: 2.244; inducer: 2.555; trim: 65
              F1-62 - exducer: 2.441; inducer: 2.795; trim: 76
              F1-65 - exducer: 2.559; inducer: 2.920; trim: 77
              F1-68 - exducer: 2.677; inducer: 3.111; trim: 74

              T4 Turbines
              "O-trim" - inducer: 2.922"; exducer: 2.296"; trim: 62
              "P-trim" - inducer: 2.922"; exducer: 2.544"; trim: 76
              "Q trim" - inducer: 3.111"; exducer: 2.693"; trim: 75​
              There is a variability in what turbine is available on T3/T04E hybrid turbos and that makes a difference in response and power. You can also get away with a smaller turbine with external wastegates than internal since your wastegate (properly placed) flows a lot more and can control boost better.

              Originally posted by Hoffbrohaus View Post
              How much boost can you safely run on an M20 with a stock head gasket and arp studs?
              A lot of power and torque, well more than you're looking at. Guides like that one by kamotors aren't really great except in the most basic (kit) circumstances since boost doesn't mean much unless you compare the same setup.You can run 20psi on a 6266 and make 500whp or 20psi on a T04E like mine and make maybe* 400whp. Density is the driving factor, not pressure, so a little turbo operating way off in the lowest efficiency zones of the map producing hot air while pushing against the restriction of a small hotside isn't moving as much air as a big turbo operating much more efficiently at the same boost with a low exhaust manifold pressure. Also, needing wasted spark above 17psi. With my setup I've run 18psi of boost on an immaculate stock ignition system + MSD coil with plugs gapped way down to 0.020", and I've had misfires at anything over 14psi if the plugs were slightly worn or the rotor not spotless. I think you need wasted spark for any serious turbo build, I'd only recommend stock system for small budget builds that aren't making more than 300whp. Even with the wasted spark I'll still get an occasional misfire telling me it's time to regap and clean or replace my plugs (wasted spark wears them out more than twice as fast as a stock ignition system). I am down on low boost while the wagon is on stands, but I was maxing out my 48lb/hr injectors (over 90% duty cycle) at 18-19psi and running 20psi on the 60lb injectors for years.

              *I say maybe because at 18psi I was walking away from an auto 2018 Mustang GT, if I was making anywhere near 400whp at 18psi I'd be leaving that car for dead since my E30 is 1,000lbs lighter which more than makes up for the shift disadvantage if you have about the same power​

              IG @turbovarg
              '91 318is, M20 turbo
              [CoTM: 4-18]
              '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
              - updated 1-26

              Comment


                #8
                Originally posted by varg View Post
                That's every turbo engine and every power goal. You are always compromising between maxing out a turbo setup to make the power you want but having good response or having worse response but the same power on less boost with lower egt. That's why VGT and twin turbo setups exist, and it's why even older factory turbo cars which can be beat on safely at OEM configuration have all of that tunable overhead, but will fail if you beat on them with the spicy tune that maxes out the stock externals.
                I came from the 4cyl turbo world, where a stock evo 9 turbo could make 400whp and spool instantly. I made 500whp on a stock frame turbo on 93 octane that loaded up like stock with huge cams. For some reason, it seems the extra two cylinders make it more complicated to get a best of both worlds setup, largely because these cars weren't turbo'd from the factory, so you're dealing within a different set of criteria. I know what you mean, though.
                1990 Brilliantrot 325iS Build Thread
                1989 Zinnoberrot M3 Build Thread

                Comment


                  #9
                  Evo9 engine is modern 4valve engine which may give wrong impression about turbo sizing. Old 2valve design must use one step bigger turbo to achieve same HP.

                  Evo9 turbocharger will not make 500whp, or even 400whp in M20.

                  Comment


                    #10
                    Originally posted by AWDBOB View Post
                    I came from the 4cyl turbo world, where a stock evo 9 turbo could make 400whp and spool instantly. I made 500whp on a stock frame turbo on 93 octane that loaded up like stock with huge cams. For some reason, it seems the extra two cylinders make it more complicated to get a best of both worlds setup, largely because these cars weren't turbo'd from the factory, so you're dealing within a different set of criteria. I know what you mean, though.
                    It's the revs. I love Evos, other than the fact that these days you pay M3 money for what is essentially an ordinary mitsubishi sedan with all of the ordinary mitsubishi sedan trappings + an awesome drivetrain. The summed flow through a 4G63 head is also probably more than the flow through an M20 head, and a stock Evo 9 revs to what? 7,500rpm? With cams and springs you guys rev them to 8,000rpm+. Having an engine that flows more than an M20 and gives you an extra 1,000rpm to work with really makes the response question relative. Having good boost response from 4,000-6,700rpm with the M20 like I do is like having it at 4,800-7,500rpm with the 4G63. So my maybe 400whp is doable with a stock evo turbo and you get full boost at 3,500rpm on a 4G63, I don't know what you'd get with an M20. It flows worse and isn't the evo twin scroll from the factory? That helps.

                    Anyway. It's not that M20s are hard to size turbos for, it's that engines like a 4G63, or a 2JZ (they're similar in this way) are much easier since the capacity to rev and flow more makes the choice easier.

                    Edit:
                    Context with my turbo, described in the above posts. This is a second gear pull starting at 4300rpm going into third with a fairly slow 0.9s shift, typical of just accelerating hard but not racing. Boost rise from atmosphere to 177kPa (~11psi target right now), is 0.6s. 4,100rpm was the post shift rpm. This is with an internal wastegate on a weird bent arm and open loop boost control. Rise time could be improved a little with an external gate and closed loop since the wastegate would stay shut tight until closer to the target. Notice the boost creep to 208kPa in 2nd.
                    Click image for larger version

Name:	image.png
Views:	762
Size:	30.1 KB
ID:	10087883​

                    ​
                    Last edited by varg; 03-15-2023, 04:32 PM.

                    IG @turbovarg
                    '91 318is, M20 turbo
                    [CoTM: 4-18]
                    '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
                    - updated 1-26

                    Comment

                    Working...
                    X