I've always wondered what the back pressure was in the exhaust manifold. I know others have done some tests measuring the back pressure in the downpipe/exhaust, but I've never seen anything done between the turbine and head.

This just seems like one of those things you actually have to measure to have a good understand of what's happening with the whole system.

In the big picture, it doesn't really matter what it is. It's not like I have an option to change the A/R in my turbine housing. But none the less, I wanted to see what it was doing.

Also this isn't exactly apples to apples with a stock MSM. Mine has been machined to fit an TD05 16G turbine wheel. The Inducer is the same size but the Exducer is 5mm larger. 3" Artech all the way back.

Anyone care to guess at the results?

 

FWIW, I logged Turbine Inlet Pressure (TIP), Turbine Outlet Pressure (TOP) and Compressor Outlet Pressure (COP) along with MAP in this post on my build thread. This is an FM2 manifold with GTX2863.

https://www.miataturbo.net/build-thr...4/#post1231551

--Ian

 

Awesome. I'll have to go take a look at your findings.

 

 

yowza

talk about some backpressure. I expected significant pressure, but not this much.

very cool test, I really appreciate the vid too

 

You have to expect more exhaust manifold pressure than boost pressure and he is on 20psi.

 

I totally forgot he's on a larger snail and high boost
so ~1.4PR?

 

Less than that, you need to look at absolute pressures, not gauge pressures.

20 psi gauge is 240 kpa, 30 psi gauge (parallax error on the video, so I can't quite tell what it says exactly) is 308. That's 1.3:1. If you assume there's 10 kpa drop in the intercooler, then it's more like 1.2:1

--Ian

 

I should do this test on the Borg to compare, I can easily remove an EGT probe to do it.

 

Have you got a spare 4+bar MAP sensor (if not, they're $10 at mouser) and analog input on the ECU? If so, I recommend data logging it, the video is cool, but the graph is much more informative.

BTW, when you do this, it helps to put a needle valve in the pressure line between the manifold and the gauge/sensor and just barely crack it open. That acts as a pneumatic low-pass filter and takes a lot of of the high-frequency noise out of the signal.

--Ian

 

So that's actually really good then huh?

And YES. You should definitely test it out. I've read that EFRs can actually run at or lower thatn 1:1.

 

I'm surprised it's that good on a stock (or rather, slightly modified) Mazdaspeed turbo.

AIUI, most modern turbos will run in "overlap" where the ratio is slightly less than 1:1 when they're at their ideal efficiency point. My 2863 does it up to about 4500 RPM.

--Ian