The hadronic cascade description developed in an earlier paper is extended to the response of an idealized fine-sampling hadron calorimeter. Calorimeter response is largely determined by the transfer of energy E_e from the hadronic to the electromagnetic sector via \pi0 production. Fluctuations in this quantity produce the "constant term" in hadron calorimeter resolution. The increase of its fractional mean, f_\rm em^0= \vevE_e/E, with increasing incident energy E causes the energy dependence of the \pi/e ratio in a noncompensating calorimeter. The mean hadronic energy fraction, f_h0 = 1-f_\rm em0, was shown to scalevery nearly as a power law in E: f_h0 = (E/E_0)m-1, where E_0\approx1~;GeV for pions, and m\approx0.83. It follows that \pi/e=1-(1-h/e)(E/E_0)m-1, where electromagnetic and hadronic energy deposits are detected with efficiencies e and h, respectively. Fluctuations in these quantities, along with sampling fluctuations, are in corporated to give an overall understanding of resolution, which is different from the usual treatments in interesting ways. The conceptual framework is also extended to the response to jets and the difference between pi and p response.

All too often we rely on Monte Carlo simulations without worrying too much about basic physics. It is possible to start with a very simple calorimeter (a big cylinder) and learn the functional form of pi e by aninduction argument. Monte Carlo simulations provide sanity checks and constants. A power-law functional form describes test beam results surprisingly well. The prediction that calorimeters respond differently to protons and pions of the same energy was unexpected. The effect was later demonstrated by the CMS forward calorimeter group, using the most noncompensating calorimeter ever built. Calorimeter resolution is dominated by fluctuations in piz production and the energy deposit by neutrons. The DREAM collaboration has recently used a dual readout calorimeter to eliminate the first of these. Ultimate resolution depends on measuring neutrons on an event-by-event basis as well.

Charge-coupled devices (CCD's) have been fabricated on high-resistivity silicon. The resistivity, on the order of 10,000 $\Omega$-cm, allows for depletion depths of several hundred microns. Fully-depleted, back-illuminated operation is achieved by the application of a bias voltage to a ohmic contact on the wafer back side consisting of a thin in-situ doped polycrystalline silicon layer capped by indium tin oxide and silicon dioxide. This thin contact allows for good short wavelength response, while the relatively large depleted thickness results in good near-infrared response.

The usual quantum efficiency (QE) measurement heavily relies on a calibrated photodiode (PD) and the knowledgeof the CCD s gain. Either can introduce significant systematic errors. But reflectivity can also be used to verify QE measurements. 1 - R >_ QE, where R is the reflectivity, and over a significant wavelength range, 1 - R = QE. An unconventional reflectometer has been developed to make this measurement. R is measured in two steps, using light from the lateral monochromator port via an optical fiber. The beam intensity is measured directly with a PD, then both the PD and CCD are moved so that the optical path length is unchanged and the light reflects once from the CCD; the PD current ratio gives R. Unlike traditional schemes this approach makes only one reflection from the CCD surface. Since the reflectivity of the LBNL CCDs might be as low as 2 percent this increases the signal to noise ratio dramatically. The goal is a 1 percent accuracy. We obtain good agreement between 1 - R and the direct QE results.

Instrumentation was developed in 2004 and 2005 to measure the quantum efficiency of the Lawrence Berkeley National Lab (LBNL)total-depletion CCD's, intended for astronomy and space applications. This paper describes the basic instrument. Although it is conventional even to the parts list, there are important innovations. A xenon arc light source was chosen for its high blue/UV and low red/IR output as compared with a tungsten light. Intensity stabilization has been difficult, but sinceonly flux ratios matter this is not critical. Between the light source andan Oriel MS257 monochromator are a shutter and two filter wheels. High-bandpass and low-bandpass filter pairs isolate the 150-nm wide bands appropriate to the wavelength, thus minimizing scattered light and providing order blocking. Light from the auxiliary port enters a 20-inch optical sphere, and the 4-inch output port is at right angles to the input port. An 80 cm drift space produces near-uniform illumination on the CCD. Next to the cold CCD inside the horizontal dewar is a calibrated reference photodiode which is regulated to the PD calibration temperature, 25$^\circ$ C. The ratio ofthe CCD and in-dewar reference PD signals provides the QE measurement. Additional cross-calibration to a PD on the integrating sphere permitslower-intensity exposures.

An overview of CCD development efforts at Lawrence Berkeley National Laboratory is presented. Operation of fully-depleted, back-illuminated CCD's fabricated on high resistivity silicon is described, along with results on the use of such CCD's at ground-based observatories. Radiation damage and point-spread function measurements are described, as well as discussion of CCD fabrication technologies.