[subexp-daq] Question on JSON output from UCESB: handling of DATA12_OVERFLOW

[Håkan T Johansson]

Dear Günter,

yes, the DATA12_OVERFLOW is stored as rawdata12 (see 
eventloop/raw_data.hh) which has the overflow mark in the highest bit 
(0x8000).  The value is in the usual bits.

Cheers,
Håkan


On Thu, 18 Apr 2024, Weber, Guenter Dr. wrote:

> 
> Dear friends,
> 
> 
> were found some unexpected (at least to us) behaviour when looking at the
> JSON output from UCESB.
> 
> 
> We are using the following command to generate the JSON outout:
> 
> 
> ../../ucesb/jena_polarimeter/jena_polarimeter stream://localhost:8003
> --ntuple=UNPACK,STRUCT,- | ../../ucesb/hbook/struct_writer -
> --dump=compact_json
> 
> In our DAQ, there are sitting CAEN V785 modules. The *.spec file for these
> modules looks like this:
> 
> 
> #define VME_CAEN_V792 VME_CAEN_V775
> #define VME_CAEN_V785 VME_CAEN_V775
> 
> VME_CAEN_V775(geom,
>           crate)
> {
>   MEMBER(DATA12_OVERFLOW data[32] ZERO_SUPPRESS);
> 
>   UINT32 header NOENCODE
>     {
>       // 0_7: undefined;
>       8_13:  count;
>       16_23: crate = MATCH(crate);
>       24_26: 0b010;
>       27_31: geom = MATCH(geom);
>     }
> 
>   list(0<=index<header.count)
>     {
>       UINT32 ch_data NOENCODE
>     {
>       0_11:  value;
> 
>       12:    overflow;
>       13:    underflow;
>       14:    valid;
> 
>       // 15: undefined;
> 
>       16_20: channel;
> 
>       24_26: 0b000;
>       27_31: geom = CHECK(geom);
> 
>       ENCODE(data[channel],(value=value,overflow=overflow));
>     }
>     }
> 
>   UINT32 eob
>     {
>       0_23:  event_number;
>       24_26: 0b100;
>       27_31: geom = CHECK(geom);
>       // NOENCODE;
>     }
> }
> 
> Now comes the strange part: from time to time we get values greater than
> 2^12 in the arrays belonging to the 12-bit ADC data (the values are than
> usually in the 36xxx ballpark). We assume that these are the overflow events
> and for some reason they are translated to values of 2^15 + something in the
> JSON output data.
> 
> 
> Is our interprtation correct and if yes, is this intended behaviour?
> 
> 
> 
> 
> Many thanks!
> 
> 
> 
> 
> Best greetings
> 
> Günter
> 
> 
> 
> 
> 
>