Choosing a turbo?

[Jacksontheogre]

I just want a usable power band, not some setup that I have to rev to 4k before boost kicks in.

[Nutzy]

I just want a usable power band, not some setup that I have to rev to 4k before boost kicks in.

What I have been trying to tell you is that big does not automatically correlate to lag. I have a data log that clearly shows my S256 hitting the set boost of 11psi at just under 3000rpm. 2500rpm is about 6psi. Running on E85 brings boost on about 200-300 rpm sooner again.

This 'big' turbo was only 400-500 rpm more laggy that the way-too-small GT2871R!

GT3076 is four cyl stuff. Stay away.

[Jacksontheogre]

What I have been trying to tell you is that big does not automatically correlate to lag. I have a data log that clearly shows my S256 hitting the set boost of 11psi at just under 3000rpm. 2500rpm is about 6psi. Running on E85 brings boost on about 200-300 rpm sooner again.

This 'big' turbo was only 400-500 rpm more laggy that the way-too-small GT2871R!

GT3076 is four cyl stuff. Stay away.

I see it sounds like the s256 is what I'm looking for, what size head gasket would I need to run it I'm assuming because full boost will be at 11psi I will have to lower compression?

[Nutzy]

If the wastegate is plumbed up direct, ie, no boost knobs etc, the stock boost is about 8-9psi which you may get away with on stock comp and, as stated, a very good tune.

If you are keen on swapping out the stock HG for a thicker Cometic, you'll have to work that one out I'm afraid. Cometic *should* have some advice on their gaskets and what compression ratio a thicker HG will achieve on your motor.

Gunni took another approach on the two M5x motors he just built; he milled the stock pistons to drop the CR. He pounded everything the S256 could make into the stock internals motor and got near 500WHP IIRC.

I see it sounds like the s256 is what I'm looking for

The 256 is a sweet street turbo, but it is not the only game in town. The HX35 is very close in size but the turbine is 5mm smaller. It is also generations older in technology, but pound for pound makes pretty much the same power. From most reports I've seen it is a more laggy unit than the 256 though.
GT3576 is another option with a 60mm or so exducer turbine with a 56mm compressor inducer. It can be had with ball bearing CHRA and divided twin scroll turbine housings.

Which brings me to another point; go twin scroll if you want faster spool!

[Jaxx_]

1. Learn how to read compressor maps
2. Use this tool to look at lots of turbos: http://www.squirrelpf.com/turbocalc/
3. ????
4. Profit.

[Nutzy]

Cool shit! I need to know how the compressor map below will work out on my M50TUB28, stock cams, 8.5:1. What RPM will the boost threshold be at? When will full boost be realised? How many psi will I have at 2500rpm?

Thanks

1. Learn how to read compressor maps
2. Use this tool to look at lots of turbos: http://www.squirrelpf.com/turbocalc/
3. ????
4. Profit.

[Q5Quint]

What year is it?



To use a compressor map: figure out what flow your engine has at what rpm, and what psi of boost you want. The compressor map will tell you if you are operating in the efficiency range of the compressor. Some have rpm instead of flow on them which makes it a little more helpful.

If your size engine is flowing 250 cubic feet a minute of air at 5000rpm- and if you want 14psi of boost at 5000rpm, your engine would need to flow double that, 500cubic feet a minute at the compressor to achieve this.

500cubic feet of air at .0807lbs a cubic foot standard temp etc is 40.35lbs if you want to put a mark on your map there.

What the compressor map does not tell you is the cfm through your exhaust turbine.

Your turbine would also need to be getting rid of 500cfm of exhaust through your turbine, while also spinning fast enough to spin your compressor wheel at the correct rpm to achieve 500 cubic feet a minute.

There are programs that do this for you based on engine size.





That program gave me about 12.17kg a minute for a m20- converted to about 27lbs again if you want to use your chart. Remember my back-asswords calc above gave us about 40.

300hp = a weeny turbo, but the throttle response is instant. Anything more than 1/4 throttle = boost. If you try to go bigger you will lose throttle response, but have more power.

The 'lag' is usually the time it takes from when you punch it to the turbo giving full boost. The little t3 super 60 has about 3psi as soon as you punch it at 2500 and it rockets up to 14 along with the rpm. The lag is not noticeable. In my mercedes 300sd if you punch it under 2500 rpm.... nothing happens- you go check your calender and go take a shit, you call your mom and breakup with your girlfriend and get over your herion addiction- then as the car approaches 3k the turbo starts spooling and boost actually starts to move the thing. That is lag. Epic unnecessary bullshit lag.

Think about the holset on my truck. It has almost instant 5psi at any rpm... because it is attached to a 6 liter diesel motor and properly sized. The cummins redlines at 3000 rpm.

A 6 liter at 3000rpm flows about as much as a 3 liter motor at 6000rpm, it just takes a 2x the rpms to spool. The cummins idling at 1200 is a m20 at 3000rpm. What happens on a cummins at 1200rpm when you push the pedal down is the same thing on a m20 with a hx35 when you push the pedal down at 3000rpm, flow wise. IE you start to haul ass.

Granted we are ignoring the whole diesel 17:1 and m20 8.8:1 compression ration thing but you get the general idea.

If you want 400ftlbs of torque, get a holset (or something similarly sized turbine and compressor). The tiny turbo was fun as hell to scooty booty around and would probably be more fun at a autocross than a holset, but if you are trying to blow the panties off anyone.... you will want the 400ft lbs of torque.

Edit: just re-read and it says you arnt trying to go crazy. 400hp should be easily achievable on a m50 and the holset (or similar) will give you room to expand later if you really want to. If you just want 8psi all day every day you might want something smaller, but once you taste 8 you will want 14 tomorrow, so might as well plan for it. If you already have the clutch go for it.

[Q5Quint]

Hate to double post but this is the next convo:

Here is what boost looks like at 458whp and 391ft/lbs on a properly sized turbo setup.

Full 30psi of boost by about 3500 rpm.




Dyno.






Now look at what car that is and realize that is just 1.8 liters with a borg warner EFR6758 turbo.



Interesting. Looks like if you are shooting for around 40/lbs a minute of flow you are well within this compressor efficiency.

Here is a holset map- just cheat and double the rpms for a 3 liter motor.

Again, dont forget to double the rpms. The slightly smaller hx35 is probably better than the hx40.

Notice how the efr turbo goes out of the range at 49lb, but the holset keeps going. forever. At a 3x pressure it is 70lbs- almost 800hp at like 40psi is possible. Diesels are neat. Their turbos are neat-er.

Size the turbine for spool time.

Here is my final image- dark blue is a .83 and light blue is a .64 a/r turbine on the exhaust side of a borg warner on a scoobyroo. See how proper sizing is important? You can pick up almost 600rpm quicker spool with the smaller exhaust housing because you dont actually need a giant .83 on a tiny ass subaru engine.






This car makes stupid power levels with a stupid 2.5 liter engine, and you can too!




Holset hx40 or 35 for cost+ 800hp ability, something smaller if you want to beat that damn miata.

Edit: links.

http://www.miataturbo.net/build-threads-57/soviet-russia-car-build-you-61522/page75/

http://www.miataturbo.net/build-threads-57/soviet-russia-car-build-you-61522/

http://www.miataturbo.net/dynos-timesheets-21/soviet-era-car-space-age-turbo-rocket-fuel-75011/

http://blog.perrinperformance.com/author/jeff-perrin/page/12/

Crap one more:

PS the reason I found the scoobyroo is they have a similarly sized engine, so the graphs would correlate.