Rigol 1052E - prevodniky

[RV]

Jojo ja to domerim...

Nastaveni bylo takove takze tech 400mV by melo sedet:
********************************************************
Analog Ch  State   Scale    Position   Coupling  BW Limit  Invert
CH1        On     50.0mV/   -2.00mV    AC        Off       Off

Analog Ch  Impedance   Probe
CH1        1M Ohm      1X

Time    Time Ref    Main Scale    Delay
Main    Center      2.000us/      0.000000s

Trigger  Source      Slope    Mode      Coupling     Level    Holdoff
Edge     CH1         Rising   Auto      DC           2.00mV   500ns

Acquisition    Sampling    Memory Depth    Sample Rate
Normal         Realtime    Normal          100.0MSa
*********************************************************
Nicmene jsem si vcera nevsiml (ted na to koukam), ze mi rigol na 
frekvenci 50kHz snizil samplovaci frekvenci na 100Msps (mozna kvuli pameti).
Uvidim dneska to jeste prozkoumam a domerim ty frekvence.


Radek Vicek


Dne 28.12.2010 12:54, balu na home napsal(a):
> tak narychlo som spracoval pripad 50kHz podla matematiky z
> predchadzajuceho mailu a vysiel fit:
>
> fittedmodel1 =
>
> General model:
> fittedmodel1(x) = a*sin(2*pi*f * x)+b*cos(2*pi*f * x)+c
> Coefficients (with 95% confidence bounds):
> a = 0.1517 (0.1516, 0.1517)
> b = 0.00558 (0.005549, 0.005611)
> c = 0.001074 (0.001052, 0.001097)
> f = 4.977e+004 (4.977e+004, 4.977e+004)
>
> a nasledne efektivny pocet bitov
>
>  >> ENOB = 8 - log2(std(error)/(0.4/256/sqrt(12)))
>
> ENOB =
>
> 6.3169
>
>
> Uvazoval som full scale 8 dielikov * 50mV = 400mV (prosim potvrdit)
>
> 6.3 bitu z 8 pri tak nizkej frekvencii (50kHz) by mohlo byt lepsie. Tie
> vyssie frekvencie budu urcite zaujimavejsie, spracujem to vecer.
> Mozeme skusit este jeden experiment. Ak je efektivny pocet bitov nizky
> koli sumu tak sa domnievam ze by mal byt konstantny vzhladom na
> frekvenciu signalu. Ak by sme ale mali k dispozicii merania na viacerych
> frekvenciach (povedzme 50k, 1M, 2M, 5M, 10M, 20M, 50M, 100M) tak mozeme
> skusit odhadnut jitter vzorkovacich hodin. Ak bude enob degradovat s
> frekvenciou signalu je problem prave s hodinami prevodnika.
> b.