All posts by Jeff

More BMW Turbocharger Inlet Flow Testing

Over the weekend I received a shipment of BMW N54 engine turbocharger inlet pipes for flow bench testing.  I’ve tested a couple other sets of these inlet pipes in the past and I was interested to see how another take on these parts would perform.

This set was supplied by FrankenTurbo, the same vendor that had shipped the stock BMW inlets to me several months ago for flow testing.

Here are a couple pictures of these FrankenTurbo inlets on the flow bench:

FrankenTurbo Inlet on Flowbench
FrankenTurbo Inlet on Flowbench

 

Second FrankenTurbo Inlet on Flowbench
Second FrankenTurbo Inlet on Flowbench

The test was performed similar to how I did in the past, affixing the inlet to the flow bench and drawing air into the pipe.

Inlet Flow Results:

The chart below is a compilation of all of the BMW turbo inlets that I have flow tested.

Chart showing BMW N54 Engine Turbo Inlet Pipe Airflow Comparison
BMW N54 Engine Turbo Inlet Pipe Airflow Comparison

In a departure from the presentation that I used in the past I am limiting the chart to airflow readings that occurred at a test depression of 28″ of H20.

There are two reasons for this; firstly 28″ of H20 is a commonly used test point for flow bench airflow tests.

The second reason is that the chart I produced previously included extrapolated data, going out to test depressions two to three times what I recorded on the bench.  Estimating out that far introduces errors, which are magnified the further one extrapolates.  To better assess how the products compare, I am limiting the results to the range that the flow bench can directly measure.

Something else to point out, I am not familiar with how these pipes attach to the N54 engine, so the labeling scheme that I used may be incorrect.  That is, I have labeled one pipe number one and the other number two.  I’m not sure that this is the proper naming convention for all of the products, in some cases I may have mislabeled pipe two as pipe one.  Keep that in mind as you review the chart data and compare airflow readings between the pipes.

Here’s an additional chart showing the FrankenTurbo inlets with the adapters added for coupling them to the turbochargers.  Airflow decreases with the addition of the adpaters.

ft_inlet_adapter

TorqByte Tuning

Slowly I’ve been trying to get up to speed on the TorqByte water-methanol injection controller and the TorqTune software application that gives access to the functionality of the controller.  The software has a good deal of capability, more than I am accustomed to from a WMI system, and thus is taking some time to get familiar with.

Up to now I have just been injecting a small amount of liquid and not given much concern to how it is being injected since my goal has been to keep IAT’s in check during dyno pulls.  Looking at some of the recent data (below) I decided to start tweaking the software settings, primarily the pump duty cycle table that drives the injection rate.

torqbyte_wmi_results_f21_rev2

I’ve chosen to try and up the WMI flow rate slightly around the point where peak boost pressure is being reached and also to slightly pull back on the injection rate beyond 5000 rpm.

The pump duty cycle table is straight forward to modify (below).

torqbyte_f21_rev2

The engine speed range is adjustable on the main page and the setting updates on the table, shown above the maximum engine speed on the table is 7500 rpm.  I dropped this down to 7000 rpm since I rarely get much past 6500 rpm.  This changed the engine speed increments from 500 rpm intervals to 400 rpm intervals.

I then went into the table and updated the duty cycle at the engine speeds of interest, around 3400 rpm and past 5000.

This configuration is saved to a unique file and then can be loaded into the TorqByte controller via a USB cable.