Lepton isolation in Delphes

[Tathagata Ghosh]

Dear Delphes experts,

I am dealing with two issues regarding lepton (e, mu) isolation in Delphes. I am using Delphes v3.3.0 within MadGraph pipeline.

I am running a heavy h2 decaying to h1 Z, where the Z boson decay to a l+l- pair (l = e, mu). My h2 has a mass of 2 TeV. So, the Z bosons coming from h2 decays are highly boosted, and the leptons coming from Z will be in a relatively narrow cone. By default, Delphes uses a cone of radius 0.5 for lepton isolation. I changed that to 0.4. However, I found that I am loosing 90% of my di-lepton events, when cross-checked with the parton level results.

So, I went through few previous tickets posted in this forum regarding lepton isolation in Delphes. In a number of those tickets you advised that to deal with similar situations one needs to set 'PTRatioMax' to a 9999 within lepton isolation module to obtain a high efficiency in leptons. However, when I set "PTRatioMax" to 9999, Delphes is not identifying any leptons at all. All the leptons are tagged as jets. This is my first issue. I am attaching my Delphes card with the ticket. Please let me know if you think that I have done something wrong here.

I talked to some of my CMS friends. They told me that to deal with two high pT leptons in a narrow cone what they do is to find a second high-pt electron or muon in the isolation cone and subtract it off from the calculation of isolation. In this way one can still use the large isolation cone size to minimize the jet-->lepton fake rate. So, my second question is that can I execute the same within Delphes in some way?

Thanks in advance,
Tathagata