solvent?

water (1.5 M solution of HF)

is fluorine added in the water in your country? I presume you use regular H2O and not D2O.

According to my recent experience of 19F-NMR on a krytox fluoropolymer and following a publication, I propose:

-86ppm: CF3

-129ppm: CF2 of alkyl (or CF -O: -132 ppm) or solvent impurities

-169ppm: CF from aromatics (hexafluorobenzene solvent for example)

Reference: http://onlinelibrary.wiley.com/doi/10.1002/mrc.2763/abstract;jsessionid=BB22FDFF0AAB4547F94B089B9D8F046C.f04t03

Hope it will help you!

Thank you Gildas, but there is no chance of presence of organic molecules in the solution, it is just concentrated HF and vanadate in water, but the fluoride does not coordinate to vanadium.

Lukáš, I am afraid you have impurities in water (or concentrated HF). Did you made a control 19F spectrum to check it ?

I don't know exactly what you made. May be vanadate was able to coordinate and concentrate impurities.

Dear Lukáš, it seems that signal at -86 could be attributed to the nonhydrated (or low hydrated, or coordinated) fluoride ion (cation is [VO2(H20)4]+). But I am not sure completely. Thus, shifts of F- is ranging from -72 (CH3CN) to -86 (THF). The shift at -129 should can be attributed to fully hydrated F- (KF(aq) shows shift about -125). The shift at 169 must be HF. Please, see several references:

http://pubs.acs.org/doi/abs/10.1021/ja01012a009?journalCode=jacsat

http://chemnmr.colorado.edu/manuals/19F_NMR_Reference_Standards.pdf

Agree with above - also remember Hf can start to etch certain types on NMR tubes for instance I have observed SiF6 2- and Al and Na F type species in these chemical shift ranges which I confirmed using Si Al and Na Nmr experiments 

At -129 to -132 ppm, a signal of SiF6 2- (etched from the NMR tude  by acid fluoride-containng solutions)  is observed. You can identify it due to the presence of satellites 19F--129Si (J = 110 Hz).

As pointed out by Gayane, the -129 ppm signal is due to SiF62- . The -169 ppm signal is very likely from HF/F-  (H-bonded species) , and the -86 ppm signal is, as Dmitrii suggested, probably from [VO2(H2O)4]+F-. Note that the [VO2(H2O)4]+ solution is said to be light yellow in color. To minimize/eliminate the SiF62- signal, use PTFE HF acid-safe NMR tubes and caps. Good luck in your research.