OWG meeting 9-Aug-2007

Agenda

1. Review minutes from the 19-Jul-2007 meeting
2. Online Action Items
3. Role of JLab Electronics group in Hall D online - Chris Cuevas
4. Hall D Online design goals and requirements - Elliott Wolin
5. Possible future meeting topics

Time/Location

10:00am Thu 9-Aug-2007 Cebaf Center F326

Announcements

• Please look at Online Design Goals, Online Major Milestones, and Online Work Breakdown and send me comments

Next Meeting

• TBD (some of us will be out of town the first week of Sep, and OECM review is second week of Sep)

New Action Items from this Meeting

• Determine requirements for and set up CODA-based Electronics group test stands - Elliott W, Chris C, Dave A
• Electronics group needs specifications from DAQ group to continue with board development - Dave A
• Milestones from Electronics group (board development, etc) - Chris C

Minutes

In attendance: Elliott W, Carl T, David L, Dave A, Vardan G, Simon T, Elke A, Elton S, Mark I, Chris C, Fernando B, Mark T, Tanest C, Bill G, Jeff W, Armenak S, Ben R, Ed J

Chris started out with a presentation on the Electronics group:

• Mission is to design/develop/support fast electronics for the entire lab.
• History goes back to the early days of the lab.
• Group has grown over the years and now has 9 people, including engineers and technicians.
• Will need to work more closely with DAQ group for the upgrade.
• Capabilities include circuit simulation, CAD, modeling, FPGA development, prototyping, assembly, testing, repair, consultation, thermal imaging, field support, etc.
• Since no more electronics pool, role here is to perform "triage" concerning whether something is broken, needs fix in-house, must be sent out, etc.
• Only minor repairs done in-house, most items sent out for repair.
• Have built and deployed many module types, of order a hundred of a single module type in some cases.
• Recent work: F1TDC, FADC, energy sum modules, VXS crate sum modules, etc. Working towards full crate sum in VXS.
  • Ed J. (DAQ group) and Ben R. are working on energy sum modules.
  • VXS not particularly popular now, but future looks solid (see [1])
• In collaboration with DAQ group:
  • testing multi-crate energy sum.
  • trigger supervisor (TS) and trigger interface (TI) modules.
  • clock/timing/sync system.
  • L1 trigger processor.
  • on-board diagnostics for above boards.
  • other custom module design and development
• Will test large systems with CODA, but will use stand-alone stations for single-board testing.
• Will help halls develop their own testing systems.
• Full crate testing will be done by Electronics group, and will help halls do the same.
• Most installation will be done by the halls, some help from Electronics group. Will NOT do major installation.
• Some other boards will be needed, but much can be done with e.g. CAEN FPGA boards that used to need custom modules, and there are a lot of commercial devices available these days.
• Need specifications from DAQ group now to continue with development of some boards.
• Will need CODA-based electronics test stands.

Next we discussed Elliott's first pass at DAQ design goals and major milestones. We did not have time to discuss Online/Controls design goals and major milestones. Many suggestions were made, and Elliott will incorporate them in the design goal and major milestone documents. Elliott will eventually create a complete work-breakdown (Gantt chart) for the DAQ/Online/Controls project.

Elke and Elton emphasized it is important to distinguish between true design goals, i.e. things absolutely needed, and desiderata, things that would be nice to have but are not required. An example of this is the ability to do multiple DAQ runs in parallel. We should not plan on a super-sophisticated trigger system to accomplish this if a simpler system without this capability will suffice for production running.

Elke further reminded us that both high-luminosity capability and the L3 farm are NOT part of the project. These will likely be needed eventually, so we need to design and build with this in mind, but items related to these two are not requirements. In other words, systems should be upgradable for high luminosity running, but are not CD-4 deliverables. Note that an event monitoring facility/prototype L3 farm IS part of the project.

Carl pointed out it is very important for designers to know what the upgraded system will look like, as this will influence critical design decisions, particularly for software. Elton pointed out that we of course will not wholesale replace lots of hardware for high-luminosity running, so most of it must meet high-luminosity specifications anyway.

I note that is somewhat of an art to know how far beyond the stated requirements one should design to, given that requirements change and new functionality is often requested. Designing too close to the requirements may result in an inflexible and hard to change system, whereas designing too far beyond requirements may waste time and manpower on unneeded features.

Finally, Elton suggested Elliott create a table of DAQ capabilities vs. when they are needed (in terms of number of crates, data and trigger rates, time needed, etc.) as part of the major milestone document.