I2773-4.4 en
167
Somat eDAQ
lite
Application Note: Guidelines for Setting the DRAM Buffering Option
If the total DRAM allocation for all Burst History channels for a single burst record is 500 kB or less, selecting the DRAM option
is advised for maximum throughput efficiency. If the total DRAM allocation is 4 MB or more, then selecting the DRAM option
is not advised and can cause a DeviceOverFlow error when the eDAQ
lite
attempts to copy the DRAM buffers to the PC Card
memory. For the gray area in the 500 kB to 4 MB range, first try not using the DRAM buffers.
Examples:
•
Do not use DRAM buffering: One Burst History DataMode defined with 16 channels of 16-bit integer data sampled at
1000 Hz for 600 seconds requires 9200000 bytes (16*2*1000*600) per burst record.
•
Use DRAM buffering: One Burst History DataMode defined with 16 channels of 16-bit integer data sampled at 500 Hz
for 20 seconds requires 320000 bytes (16*2*500*20) per burst record.
•
Try not using DRAM buffering: One Burst History DataMode defined with 16 channels of 16-bit integer data sampled
at 500 Hz for 10 seconds requires 320000 bytes (16*2*500*20) per burst record and a second burst DataMode defined
with 8 channels of 32-bit float data sampled at 2000 Hz for 10 seconds requires 640000 bytes (8*4*2000*10) per burst
record for a total of 960000 bytes.
8.3.3 Event Slice
The Event Slice DataMode stores a set of master channels and a set of slave channels
in the output data file. The set of master channels provides a sequence of events,
which are defined as changes in the state of any master input channel. For each event,
the eDAQ
lite
stores data samples for all channels (masters and slaves) in the output
data file.
Master Input Channels
Select the desired set of master input channels using CTRL to select more than one
channel. The master input channels can be any data type.
Slave Input Channels
Select the desired set of slave input channels using CTRL to select more than one
channel. The slave input channels can be any data type.
Data Type
For 32-bit float input channels, select one of three available formats for data storage
and conversion.
NOTE
Using the integer data types requires that valid full-scale minimum and maximum
values are defined for each input channel selected for the DataMode.
Data Type Description
32-bit float No conversion is necessary. Use the 32-bit float mode for computed
channels where full-scale estimates are uncertain or unknown.
16-bit integer Maintain the resolution of both the 12-bit and 16-bit A/D converters.
8-bit integer Lose significant resolution in both the 12-bit and 16-bit A/D converters.
Use the 8-bit integer mode only for a rough picture of channel
behavior.
HBM: public
To Next Page
Loading...