Okay guys, I think I've cracked it for the HE221 and TD04HL.
Here's what I did.
1. I downloaded a QSB4.5 CAD model a while back. This had an outside surface model of an HE221 turbo. I'm assuming this turbo has the 6cm housing.
2. I cut the turbo CAD model parrallel to the exhaust flange at shaft height.
3. I assumed the walls of the scroll were about 4mm thick, so I offset the outer edges and made a shape to resemble the port.
4. I adjusted the width of this port to give an area of 6cm^2.
5. I measured on the computer the distance from the port shape I'd just created to the centreline of where the shaft would be.
6. Divided A by R and converted to inches.
http://www.aulro.com/afvb/attachment...1&d=1381892532
Now it's not 100% because I've had to assume two things.
1. The shape of the port and wall thickness I had to make up.
2. That the model I had was a 6cm housing.
The results come out with 6cm area and 4.9cm radius. A/R is 1.22cm or 0.48 inches.
This was the second attempt. The first one I used a 5mm wall thickness and due to the changes in port shape that gave a result of 0.51 inch A/R.
So it's not exact, but I'm pretty damn sure I'm within 10%.
Now here is the interesting part that I'm sure turbo makers really hate me doing.
I've collected all the turbine flow figures from the garrett turbine maps and arranged them all in a table with exducer size, A/R and max corrected flow.
I've thrown the Borg Warner EFR turbine maps for the EFR6255 and EFR6258 in there for good measure.
I've found a quite common relationship across all of them.
A/R doesn't change the flow linearly, it's a square root relationship. So if you half the A/R you change the max flow by about the square root of half or 0.7.
If I divide the exducer flow area by the square root of the A/R I get a consistent number across all turbines of similar size.
As the turbines get much larger, this number drifts a bit smaller. But the trend is very good and easy to follow.
Upshot. I can predict max flow and the rough shape of a turbine map from the turbine exducer and the housing A/R.
So for the HE221 or TD04HL with a 6cm housing (same thing) we have an exducer of 45.6mm and an A/R of 0.5.
The max corrected flow is expected to be ~14.7 lb/min which is just a fraction bigger than the Garrett GT22 (14.2lb/min), significantly bigger than the 0.49 A/R T25 I'm running now (approx 13.3 lb/min). But approx 10% smaller than the 0.64 A/R T25/T28 I had run for a while.
In short, it is the size I hoped it would be.
This also lets me work out the max power and hence max boost/flow this turbine can support.
Using EGT of 700C (1300F), backpressure of 120kPa (3psi) and a pressure ratio of 3 we have a real turbine flow of ~28.4 lb/min, drive pressure of 37.5psi and turbine power of 41.5kw.
The temperature drop across the turbine is 192C.
This 41.5kw of turbine power can drive 29psi of boost to 4000rpm on a 4BD1T. It reaches right to the top and right of the HE221 and TD04HL-19T compressor maps, which is a good sign that the numbers are close to correct.
This amount of boost could deliver approximately 207kw at 4000rpm on a 4BD1T.
