I grabbed the files of electron candidate tracks from Matt Walker which he used for his 10 Electron E/p from pp500 L2W events from the st_W sample production, and tried to make a preliminary plot of the ratio of negative to positives vs. pt for east and west. His TTrees saved pretty much all global tracks with momentum above 1 GeV/c which have a primary track partner, so I merely implemented only a few additional basic cuts. Note that the high-pt data here probably has a significant number of electrons, so it is more accurate to call this -/+ than specifically h-/h+ as these are not all hadrons. However, I believe this is moot in the discussion of momentum distortions.

Please keep in mind that this production used preliminary TPC distortion corrections, most of which are east-west symmetric. The TPC Twist correction is antisymmetric at any given phi, but should be roughly symmetric for phi-integrated studies. Smaller order east-west differences may come from various dead TPC regions and anode high voltages, and perhaps T0s and alignment. This must be considered a rough first look.

Anyhow, my additional cuts to the tracks are:

1. |eta +/- 0.6| < 0.4 to define east vs. west (this may not be very stringent)
2. nFitPoints >= 25
3. flag < 500

The result is the following plot of the ratio versus momentum (note that this is not strictly p_T, but this should not be an important point in comparing east vs. west):

What we see from this preliminary work is that there is no strong east vs. west difference in the ratio (perhaps my cut for east and west is not sufficient). Compare this to results from Run 8 (see for example pp data, and dAu data) which show clear differences by 5 GeV/c (~0.7 on my plot), getting worse at higher p_T. The st_W sample is certainly far from minimum bias, though I'm not sure whether the trigger bias can explain the ratio staying close to 1.0 (even above it) out to 6 GeV/c (~0.78 on my plot).

Fig. 6 on 10 Electron E/p from pp500 L2W events indicates that the positively charged tracks are being assigned momenta which are too high (and thus E/p is lower for them) by a few percent. If that is the case, it may be that my ratios are smaller than they should be (the denominator in the -/+ ratio at each reconstructed momentum bin is too large since it's really from a lower momentum), and the ratios should be closer to 1.0 at high pp. However, Matt has not divided up east and west on his plot, so it is not a clear comment on differences between east and west TPC.

This is only RFF data. We will need to process Run 9 pp200 data to compare RFF to FF, taken during the last 6 weeks of operations.

-Gene