In some cases (e.g. some organic systems in biology field), lower voltage TEM and STEM cannot significantly reduce beam damage since knock-on damage is not the main mechanism for such materials. In these cases, the electron dose is the most important factor on specimen damage.

The typical method for obtaining low dose is to change the parameters of the electron gun. However, this modifies the electron optics in the column. For a STEM with a C_s corrector, this also causes a misalignment of the corrector. Unfortunately, it is hard to re-align the column as well as the corrector under low-dose conditions since the auto tuning software or even manual adjustment normally needs high signal-to-noise images and/or diffractograms to converge precisely.

Note that artifacts can exist in the low dose imaging:
        i) A streaking artifact parallel to the scanning direction. Such streaking artifact is caused by a slow reaction time of the photomultiplier/readout electronics. This streaking effect can be significantly reduced by increasing the dwell time (scanning slower) or increase the electron dose.
        ii) An artifact due to a random horizontal offset of scan lines.

Typical low dose STEM imaging can be performed at a dose of ~15-20 e^-/Ų with a probe current of ~1 pA and a scan speed of 1.0 µs per pixel. In this case, the signal to noise ratio (SNR)of recorded images is so low that the naked eyes are not able to recognize any structural information. However, useful high magnification images at an atomic resolution can still be obtained by Fourier-filtering method:
         Step i) Convert this image to Fourier-transformed pattern (e.g. with FFT).
         Step ii) Use a mask to select the spots.
         Step iii) Reversely convert the Fourier-transformed pattern back to real-space image.

Note that a good method to do low-dose STEM imaging is to use a proper fast shutter to move the beam on/off the specimen to control the dose and to avoid any change in the electron optics.