ETTC'2003 - SEE

ETTC'2003 - SEE ETTC'2003 - SEE

web1.see.asso.fr
from web1.see.asso.fr More from this publisher
21.04.2013 Views

For the purposes of this paper it is interesting to note that with digital filters, four more parameters can be removed from the specification table, as they are typically an order of magnitude less with digital filters. These parameters include: o Passband-ripple o Fc accuracy o Fc drift o Time or Phase Delay variation 5. CONCLUSIONS There are many possibilities for the next generation of analog signal conditioning in airborne data acquisition systems. These include: • 16-bit resolution on ADC’s • >14-bit ENOB • Temperature compensation of gain and offset • Optimum ADC’s for each application • Optimum filtering so as to minimize data rates (not just a limited choice) • Optimum gain and offset settings (not just a limited choice) Although the technology behind the next generation signal conditioning may be new, the good news is that there are considerably less specifications for the system designer to be concerned with. This paper has outlined what parameters may be most important when specifying the next generation signal conditioning circuits. In particular a rigorous definition of: • Total DC error (including excitation) • Effective number of bits (as a function of gain and frequency) and insistence on digital filtering means that different modules from different vendors can be compared more easily. The net effect of all of this is to make the following specifications superfluous: Linearity (DNL, INL), Quantization noise, Jitter, Jitter noise, Cross-talk, Channel-tochannel matching, Resolution, Missing codes, THD, SNR, SFDR, SINAD, Gain error, Gain drift, Offset error, Offset drift, Excitation error, Excitation drift, Ripple in passband, Filter cut-off drift, Filter delay variation. The benefits of this approach are clear.

BACK Building bridges between networks Diarmuid Corry Director of Software Engineering corry@acracontrol.com ACRA CONTROL Ltd. ACRA CONTROL Inc. Landscape House 44145 Airport View Drive Landscape Road Hollywood Dublin 14 MD 20636 Ireland USA Abstract The technologies used for flight test are evolving. Trusted standards like CAIS and IRIG106 PCM are giving way to new “trusted standards” (and proven technologies/protocols) found in telecommunication and networking such as Ethernet, fibre channel, TCP/IP, UDP, ATM and so on. Currently there is Eur 100Ms+ invested in CAIS and IRIG compliant equipment in the world. A key challenge in this evolution is to provide a reliable solution that allows the FTI engineer to immediately take advantage of these advanced technologies while protecting prior investment in equipment, knowledge, and resources during this transition. This paper presents an analysis of how to protect existing assets while still leveraging the power of the latest technologies. It looks at the characteristics of a “bridge” system, and suggests solutions for merging and linking data from and to different transmission protocols using data synchronization and deterministic data management cycles. Keywords Flight test instrumentation, Networks, CAIS, Ethernet, Gateway Overview Over time the world of flight test instrumentation has developed a method of operation that addresses the complex problem of acquiring co-related data in a hostile environment. Figure 1 shows a typical complex flight test data acquisition system. It uses several protocols that are common within the flight test community (IRIG106 PCM, CAIS, MIL-STD-1553 etc.) This configuration and its associated protocols will have been developed over months for a specific test sequence. It provides synchronous and time-tagged data reliably; with rigorously defined timing sequences and using no microprocessors. Outside of the flight test industry the commercial world has been solving a different set of problems. The concern here is to transfer large amounts of data over illdefined (and often dynamic) network infrastructures reliably. The time taken is not important (within reason) and some data loss is tolerable. Compared to the flight test world, the commercial world is huge - $100B/annum compared to $100m/annum. Competition in this world has driven the cost of

For the purposes of this paper it is interesting to note that with digital filters, four more<br />

parameters can be removed from the specification table, as they are typically an order of<br />

magnitude less with digital filters.<br />

These parameters include:<br />

o Passband-ripple<br />

o Fc accuracy<br />

o Fc drift<br />

o Time or Phase Delay variation<br />

5. CONCLUSIONS<br />

There are many possibilities for the next generation of analog signal conditioning in airborne<br />

data acquisition systems. These include:<br />

• 16-bit resolution on ADC’s<br />

• >14-bit ENOB<br />

• Temperature compensation of gain and offset<br />

• Optimum ADC’s for each application<br />

• Optimum filtering so as to minimize data rates (not just a limited choice)<br />

• Optimum gain and offset settings (not just a limited choice)<br />

Although the technology behind the next generation signal conditioning may be new, the<br />

good news is that there are considerably less specifications for the system designer to be<br />

concerned with.<br />

This paper has outlined what parameters may be most important when specifying the next<br />

generation signal conditioning circuits. In particular a rigorous definition of:<br />

• Total DC error (including excitation)<br />

• Effective number of bits (as a function of gain and frequency)<br />

and insistence on digital filtering means that different modules from different vendors can be<br />

compared more easily.<br />

The net effect of all of this is to make the following specifications superfluous:<br />

Linearity (DNL, INL), Quantization noise, Jitter, Jitter noise, Cross-talk, Channel-tochannel<br />

matching, Resolution, Missing codes, THD, SNR, SFDR, SINAD, Gain error,<br />

Gain drift, Offset error, Offset drift, Excitation error, Excitation drift, Ripple in passband,<br />

Filter cut-off drift, Filter delay variation.<br />

The benefits of this approach are clear.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!