As mentioned above, check your lamp. Also make sure the flow cell is clean and the quartz isn't solarized from the UV light

In addition to the suggestions above, what solvent system are you using? If you are running solvents such as ethyl acetate, benzene, toluene, dichloromethan, or other light-absorbing solvents (used for normal-phase HPLC or non-aqueous reverse phase), they could absorb light and mask the spectrum of your compound(s). Also, if using transparent solvents, make sure they aren't contaminated with other solvents or compounds since the absorption may block the light to the detector as well.

thank you.

I checked the detector by applying toluene and methanol. I saw the pick of  toluene in the spectrum. for unknown samples, I use CH3CN, formic acid and water as mobile phase. but I do not any pick in the spectrum.

Summary so far:

• Running methanol and injecting toluene, a spectrum of toluene is seen- suggests detector is ok
• Running your mobile phase, unknown compounds are not seen

As it appears the detector is working, items to check (in no particular order):

• Solvent contamination- if there is absorbing material in your mobile phase, it can block light enough to prevent measuring a spectrum. Suggest measuring the absorbance of your mobile phase on a UV-vis spectrometer.
• The sample has no chromophores- again, test on a UV-vis spectrometer. It is possible the unknown doesn't contain the expected compound(s)
• The compounds haven't eluted from the column (so can't be detected at the other end of the column), or elute in a broad band such that they are not easily detected. Suggest running a gradient to see if the compounds elute in a greater concentration of CH3CN than expected. If using a silica-based C18 column, try washing with dichloromethane to see if the compound elutes (this solvent absorbs UV light, but washes lipophilic compounds from C18. Polymeric reverse phases generally aren't compatible with DCM).
• Compounds didn't elute- I repeat this due to a different cause; perhaps the pumping system wasn't working properly so the compounds failed to elute- could be air in the pump, or a check valve not working, ot other pumping issues.

Make sure you have your PDA setup in "scanning mode" between two wavelength ranges. If you have not turned this feature on, you will not be able to use the 3D spectral tools in your software.